hysts-bot commited on
Commit
a4e70a7
·
verified ·
1 Parent(s): eb68269

Upload folder using huggingface_hub

Browse files
0.codes.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:a4429348f4775c6d538c5e58ee1a163a1b59eaaf996ac24836d73532c710a1dd
3
- size 3766748
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3e94e9a973b6810651c86639143e85abb0d3c3eb1dd92f74afa3b98331bcaf19
3
+ size 3767324
0.metadata.json CHANGED
@@ -1,6 +1,6 @@
1
  {
2
  "passage_offset": 0,
3
- "num_passages": 5497,
4
- "num_embeddings": 941392,
5
  "embedding_offset": 0
6
  }
 
1
  {
2
  "passage_offset": 0,
3
+ "num_passages": 5498,
4
+ "num_embeddings": 941544,
5
  "embedding_offset": 0
6
  }
0.residuals.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:2e4466799327fc36992f3ce98e33fa9b1157116b1e661afb7d0beb6d3419f1bb
3
- size 60250288
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:90d5c24f24298949ba59ebdaad233d5a651a95db67814e2fdc62adfbd164dd2c
3
+ size 60260016
avg_residual.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:29f2eef965a59b050aeb83c37f166091a6431d5d1766f1513ce2d2eaff766902
3
  size 1205
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0314f57b388bde4df93b1be2c1a2e0fd6a65ea0b7b6916c3deb39b3a766592e8
3
  size 1205
buckets.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:5875515a8f9071a80c5df1fbd9140b94f6b1f1acb9c520cd3c1e716a1045f2e1
3
  size 1432
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1a098a0b5146e2e71ad84a75997f4528c76dc2d87900a7e69957b5abf130fa09
3
  size 1432
centroids.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:2d5f556e499d9f1f74250572ba9ca6b65bdef0ada73899b17e410a037dce117c
3
  size 2098342
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6edb759f5540755d1ba309ba34b3317fe3d52552f5493f14a310911f4176d5ec
3
  size 2098342
collection.json CHANGED
@@ -5495,5 +5495,6 @@
5495
  "Understanding and accurately following instructions is critical for large language models (LLMs) to be effective across diverse tasks. In this work, we rigorously examine the key factors that enable models to generalize to unseen instructions, providing insights to guide the collection of data for instruction-tuning. Through controlled experiments, inspired by the Turing-complete Markov algorithm, we demonstrate that such generalization only emerges when training data is diversified enough across semantic domains. Our findings also reveal that merely diversifying within limited domains fails to ensure robust generalization. In contrast, cross-domain data diversification, even under constrained data budgets, significantly enhances a model's adaptability. We further extend our analysis to real-world scenarios, including fine-tuning of $textbf{specialist} and textbf{generalist}$ models. In both cases, we demonstrate that 1) better performance can be achieved by increasing the diversity of an established dataset while keeping the data size constant, and 2) when scaling up the data, diversifying the semantics of instructions is more effective than simply increasing the quantity of similar data. Our research provides important insights for dataset collation, particularly when optimizing model performance by expanding training data for both specialist and generalist scenarios.",
5496
  "Our research provides important insights for dataset collation, particularly when optimizing model performance by expanding training data for both specialist and generalist scenarios. We show that careful consideration of data diversification is key: training specialist models with data extending beyond their core domain leads to significant performance improvements, while generalist models benefit from diverse data mixtures that enhance their overall instruction-following capabilities across a wide range of applications. Our results highlight the critical role of strategic diversification and offer clear guidelines for improving data quality.",
5497
  "Reinforcement learning from human feedback (RLHF) has demonstrated effectiveness in aligning large language models (LLMs) with human preferences. However, token-level RLHF suffers from the credit assignment problem over long sequences, where delayed rewards make it challenging for the model to discern which actions contributed to successful outcomes. This hinders learning efficiency and slows convergence. In this paper, we propose MA-RLHF, a simple yet effective RLHF framework that incorporates macro actions -- sequences of tokens or higher-level language constructs -- into the learning process. By operating at this higher level of abstraction, our approach reduces the temporal distance between actions and rewards, facilitating faster and more accurate credit assignment. This results in more stable policy gradient estimates and enhances learning efficiency within each episode, all without increasing computational complexity during training or inference. We validate our approach through extensive experiments across various model sizes and tasks, including text summarization, dialogue generation, question answering, and program synthesis. Our method achieves substantial performance improvements over standard RLHF, with performance gains of up to 30% in text summarization and code generation, 18% in dialogue, and 8% in question answering tasks.",
5498
- "Our method achieves substantial performance improvements over standard RLHF, with performance gains of up to 30% in text summarization and code generation, 18% in dialogue, and 8% in question answering tasks. Notably, our approach reaches parity with vanilla RLHF 1.7x to 2x faster in terms of training time and continues to outperform it with further training. We will make our code and data publicly available at https://github.com/ernie-research/MA-RLHF ."
 
5499
  ]
 
5495
  "Understanding and accurately following instructions is critical for large language models (LLMs) to be effective across diverse tasks. In this work, we rigorously examine the key factors that enable models to generalize to unseen instructions, providing insights to guide the collection of data for instruction-tuning. Through controlled experiments, inspired by the Turing-complete Markov algorithm, we demonstrate that such generalization only emerges when training data is diversified enough across semantic domains. Our findings also reveal that merely diversifying within limited domains fails to ensure robust generalization. In contrast, cross-domain data diversification, even under constrained data budgets, significantly enhances a model's adaptability. We further extend our analysis to real-world scenarios, including fine-tuning of $textbf{specialist} and textbf{generalist}$ models. In both cases, we demonstrate that 1) better performance can be achieved by increasing the diversity of an established dataset while keeping the data size constant, and 2) when scaling up the data, diversifying the semantics of instructions is more effective than simply increasing the quantity of similar data. Our research provides important insights for dataset collation, particularly when optimizing model performance by expanding training data for both specialist and generalist scenarios.",
5496
  "Our research provides important insights for dataset collation, particularly when optimizing model performance by expanding training data for both specialist and generalist scenarios. We show that careful consideration of data diversification is key: training specialist models with data extending beyond their core domain leads to significant performance improvements, while generalist models benefit from diverse data mixtures that enhance their overall instruction-following capabilities across a wide range of applications. Our results highlight the critical role of strategic diversification and offer clear guidelines for improving data quality.",
5497
  "Reinforcement learning from human feedback (RLHF) has demonstrated effectiveness in aligning large language models (LLMs) with human preferences. However, token-level RLHF suffers from the credit assignment problem over long sequences, where delayed rewards make it challenging for the model to discern which actions contributed to successful outcomes. This hinders learning efficiency and slows convergence. In this paper, we propose MA-RLHF, a simple yet effective RLHF framework that incorporates macro actions -- sequences of tokens or higher-level language constructs -- into the learning process. By operating at this higher level of abstraction, our approach reduces the temporal distance between actions and rewards, facilitating faster and more accurate credit assignment. This results in more stable policy gradient estimates and enhances learning efficiency within each episode, all without increasing computational complexity during training or inference. We validate our approach through extensive experiments across various model sizes and tasks, including text summarization, dialogue generation, question answering, and program synthesis. Our method achieves substantial performance improvements over standard RLHF, with performance gains of up to 30% in text summarization and code generation, 18% in dialogue, and 8% in question answering tasks.",
5498
+ "Our method achieves substantial performance improvements over standard RLHF, with performance gains of up to 30% in text summarization and code generation, 18% in dialogue, and 8% in question answering tasks. Notably, our approach reaches parity with vanilla RLHF 1.7x to 2x faster in terms of training time and continues to outperform it with further training. We will make our code and data publicly available at https://github.com/ernie-research/MA-RLHF .",
5499
+ "Long-context language models (LCLM), characterized by their extensive context window, is becoming increasingly popular. Meanwhile, many long-context benchmarks present challenging tasks that even the most advanced LCLMs struggle to complete. However, the underlying sources of various challenging long-context tasks have seldom been studied. To bridge this gap, we conduct experiments to indicate their difficulty stems primarily from two basic issues: \"multi-matching retrieval,\" which requires the simultaneous retrieval of multiple items, and \"logic-based retrieval,\" which necessitates logical judgment within retrieval criteria. These two problems, while seemingly straightforward, actually exceed the capabilities of LCLMs because they are proven to be hyper-multi-step (demanding numerous steps to solve) in nature. This finding could explain why LLMs struggle with more advanced long-context tasks, providing a more accurate perspective for rethinking solutions for them."
5500
  ]
doclens.0.json CHANGED
@@ -1 +1 @@
1
- [206,104,226,67,200,185,221,212,206,222,88,210,228,174,155,205,172,218,148,132,212,91,163,184,205,132,213,190,212,198,230,227,159,198,122,216,175,197,118,217,219,224,69,220,197,72,204,92,169,191,191,155,175,111,218,77,207,36,195,178,123,170,231,91,209,177,146,205,151,221,217,95,199,89,153,216,154,202,167,213,104,184,176,226,200,232,53,223,73,202,220,158,174,189,165,222,158,221,211,78,205,72,214,212,215,232,175,219,110,205,192,226,34,176,209,158,222,132,179,208,98,126,205,126,172,167,227,100,224,182,208,117,217,199,195,191,169,178,158,217,152,157,163,163,204,209,146,217,150,194,217,23,125,200,221,200,212,41,176,223,207,95,178,232,68,182,173,205,198,210,139,152,147,215,196,223,83,199,123,197,119,230,223,65,216,85,210,52,210,204,179,111,138,215,83,177,219,69,212,77,182,226,99,178,207,197,87,190,222,216,85,208,211,66,220,226,221,134,175,190,170,166,157,216,135,211,132,200,67,227,195,178,221,179,205,168,186,208,127,207,139,205,67,209,223,117,195,190,202,64,218,77,199,221,189,125,216,80,148,214,143,181,98,221,194,112,219,212,138,216,87,127,187,219,122,220,97,223,221,164,205,220,100,227,101,171,188,223,89,213,137,175,172,219,77,102,234,109,197,114,205,184,221,138,175,152,163,133,227,73,224,181,174,147,144,178,147,224,211,121,123,211,219,209,210,204,80,182,215,152,118,224,216,154,220,163,204,187,133,151,162,221,94,184,224,213,72,187,148,223,195,110,182,117,224,88,202,162,220,154,151,197,227,197,74,221,205,97,221,219,103,206,184,72,191,201,192,230,164,217,170,214,118,221,218,118,215,181,224,78,183,154,190,206,102,202,74,195,96,225,228,221,99,224,104,123,203,184,211,90,209,173,199,203,99,201,176,160,207,132,231,96,195,213,173,101,190,206,61,225,232,215,178,211,227,192,148,195,213,114,212,127,162,214,213,129,220,70,176,224,209,137,232,116,228,212,75,182,207,71,111,173,189,207,96,172,213,119,222,93,195,204,211,76,179,200,184,216,107,135,202,71,118,219,123,202,210,146,225,76,227,62,214,226,98,158,202,219,169,83,156,223,125,211,213,102,219,223,72,224,178,190,204,110,228,64,214,200,193,205,213,160,201,216,134,195,128,186,213,219,187,199,216,84,202,150,211,127,230,206,204,180,183,183,167,213,224,72,209,79,195,205,153,228,133,203,210,78,190,207,195,133,225,192,197,190,201,187,222,109,138,227,103,223,209,164,190,73,226,200,205,191,189,100,226,135,198,129,185,207,229,148,209,216,58,222,117,219,92,221,147,222,219,63,216,228,165,181,157,213,197,218,158,191,186,223,204,109,172,229,91,231,204,102,192,217,204,144,222,173,209,209,95,186,195,60,212,219,209,162,183,227,216,111,193,182,222,71,183,222,122,178,219,130,228,82,225,223,69,198,232,63,193,207,216,139,223,74,164,198,224,80,148,227,138,227,199,225,86,148,230,169,210,92,166,187,202,165,196,78,202,74,229,173,216,215,213,164,205,102,206,91,212,173,198,109,183,94,206,115,224,126,229,222,57,225,191,221,218,215,44,177,217,214,218,191,221,80,198,109,196,95,192,181,214,214,123,208,191,225,219,218,108,204,150,199,206,90,227,71,207,123,199,106,131,213,93,178,208,60,132,217,149,204,80,227,218,202,94,232,175,213,115,217,114,222,200,173,184,217,222,178,226,205,192,185,197,180,193,205,211,136,221,62,184,213,89,225,71,169,216,135,198,98,214,204,88,191,154,207,213,227,93,230,133,216,189,122,224,108,229,237,218,209,214,66,153,203,219,204,222,218,107,174,214,148,169,188,210,98,203,142,167,174,224,103,223,220,148,216,199,188,175,173,130,156,97,220,111,216,115,216,70,221,210,77,173,162,216,164,216,69,203,175,172,166,199,156,184,131,201,210,98,220,133,211,116,221,82,213,88,213,128,227,219,177,190,216,180,237,217,64,215,134,221,172,220,221,70,169,221,189,226,74,229,197,209,128,190,213,106,221,198,162,184,182,219,86,201,143,229,205,146,228,106,154,209,60,216,222,74,221,215,141,168,202,82,194,213,80,217,214,204,164,199,220,149,114,140,215,150,218,108,109,211,170,206,125,183,148,193,229,210,124,188,143,160,153,199,178,75,225,105,199,83,222,112,205,75,222,205,70,104,194,91,158,220,99,215,209,219,65,124,215,158,223,234,208,101,218,200,66,188,222,149,201,189,211,51,179,208,198,160,182,152,195,141,216,112,205,207,170,128,219,118,188,223,224,80,177,218,127,196,151,159,117,186,113,216,212,134,204,90,200,231,184,222,68,146,151,158,202,173,144,206,171,215,68,188,222,89,215,117,223,80,191,209,204,105,228,67,174,210,74,208,105,151,159,183,142,163,202,214,41,147,222,227,188,208,200,157,132,195,94,191,185,214,132,201,155,166,195,220,147,219,156,214,165,220,105,223,217,100,176,225,117,207,134,166,202,218,172,200,157,183,217,87,214,181,229,217,76,231,199,139,154,213,208,208,169,218,82,219,191,139,207,222,49,122,154,176,219,110,185,112,137,231,80,208,136,189,192,208,190,192,187,180,220,186,225,218,208,102,190,124,152,83,219,199,95,198,133,235,214,96,213,217,148,168,200,32,145,207,222,198,171,208,218,104,178,230,77,214,129,218,136,139,212,233,93,221,153,221,95,155,151,211,225,60,81,217,201,108,208,234,172,177,172,193,208,72,216,224,211,44,205,82,221,224,199,198,221,137,208,81,183,105,182,106,222,180,184,146,180,205,187,192,163,219,213,164,229,218,190,147,221,193,213,182,126,228,231,97,173,218,116,161,156,207,122,193,198,155,207,103,223,192,92,153,223,170,206,177,225,40,224,106,203,208,116,203,105,197,152,221,211,84,226,111,215,154,224,221,65,215,142,169,212,184,135,207,85,209,213,76,225,229,130,203,217,186,229,140,177,116,216,219,62,209,105,164,150,207,214,211,114,148,230,118,201,134,197,124,209,163,184,164,186,189,219,106,221,220,95,141,217,87,200,224,114,220,168,212,112,229,192,106,189,183,209,212,85,187,202,219,222,223,223,212,226,79,193,128,222,209,206,197,178,152,114,222,111,164,183,189,179,214,130,149,150,212,122,160,195,118,115,204,144,163,159,140,211,208,120,222,220,125,195,174,220,80,173,204,87,149,212,82,221,119,213,139,186,227,97,215,213,136,215,68,229,218,63,224,182,221,218,94,213,90,168,202,85,168,222,121,215,213,93,221,83,220,112,197,213,213,92,210,76,216,79,179,220,94,194,99,186,181,197,214,194,138,221,91,218,172,159,219,74,153,221,201,108,213,131,217,112,214,107,209,35,224,188,219,108,191,105,157,211,126,206,148,220,126,217,164,203,129,201,211,119,204,165,209,75,205,130,204,215,206,119,178,211,218,137,193,167,199,220,182,115,207,184,189,217,178,186,159,208,230,140,194,75,173,205,64,213,167,189,110,220,195,136,137,196,134,208,120,198,182,189,176,162,183,172,172,214,200,95,212,189,193,184,195,116,167,199,110,180,217,126,193,226,142,204,66,140,222,226,218,144,218,214,140,206,66,196,224,143,121,166,221,101,154,204,216,201,221,152,225,69,226,226,180,200,181,212,128,193,206,136,219,141,208,128,202,209,105,168,178,204,211,110,145,163,193,211,69,220,147,194,223,165,212,218,124,197,229,226,132,220,156,211,80,203,140,192,94,208,102,180,215,134,152,218,106,191,204,102,211,108,182,204,206,208,90,165,197,105,186,115,213,208,186,165,186,174,213,211,164,210,62,129,201,217,170,169,217,193,167,222,174,217,167,128,182,113,159,208,85,172,219,89,219,126,218,190,65,227,208,52,225,125,160,205,135,168,186,213,212,69,124,205,67,204,110,151,180,220,205,193,106,223,80,124,166,218,194,138,148,205,88,179,183,163,227,221,198,213,87,190,200,239,81,202,211,213,115,204,105,216,83,170,201,210,95,219,219,180,216,222,228,86,231,83,227,116,223,84,207,216,115,204,109,194,171,198,183,197,86,216,144,209,125,226,221,57,215,117,199,185,152,230,174,220,61,209,99,163,218,218,134,183,212,85,208,92,190,214,81,146,220,181,224,214,212,79,171,209,217,209,100,213,72,220,223,193,72,130,203,111,214,192,197,222,201,229,204,189,162,124,221,157,218,34,204,88,170,205,76,168,210,46,196,223,73,180,85,118,200,85,221,68,160,202,201,219,181,148,207,210,208,69,210,101,215,76,217,104,188,180,224,71,199,166,203,106,216,196,182,84,183,125,128,224,89,206,88,212,212,152,205,92,210,65,166,171,200,120,228,213,83,163,131,153,180,215,228,99,179,210,82,218,95,209,94,215,206,196,110,236,219,209,193,95,172,177,218,218,140,203,228,102,160,215,102,204,192,89,168,222,56,223,155,189,213,208,220,78,204,213,168,202,169,230,191,218,203,73,222,127,208,143,221,217,217,100,160,212,208,156,223,197,205,201,223,220,144,209,90,209,139,182,218,22,221,183,219,79,216,194,224,86,208,225,190,102,200,212,162,211,83,221,106,215,113,161,197,111,216,210,92,219,124,208,185,152,203,220,230,106,182,193,185,221,79,197,79,231,192,202,113,166,211,142,195,170,179,226,222,208,228,152,133,200,89,197,180,155,215,97,201,139,205,145,217,163,213,85,223,209,200,139,217,123,160,183,201,136,210,206,105,208,132,206,208,101,134,174,112,205,181,221,215,221,213,225,116,217,80,162,186,120,183,122,214,201,97,188,209,174,229,205,120,194,203,130,228,134,228,83,201,119,218,210,55,225,64,171,174,227,107,226,95,139,190,154,171,206,139,154,219,157,224,213,206,212,79,135,186,156,217,186,128,118,184,187,213,221,86,183,217,218,210,216,155,195,188,134,172,223,125,176,167,85,187,214,217,94,213,128,217,133,192,217,60,189,122,217,117,224,95,199,74,206,89,221,205,158,196,77,228,114,188,105,210,70,209,83,188,169,214,225,184,205,127,197,115,182,174,181,238,222,223,59,229,219,112,205,115,190,128,196,210,222,176,215,194,187,112,212,90,183,207,86,212,85,178,187,138,217,80,203,96,221,198,82,207,96,164,227,208,76,214,179,209,186,207,205,221,224,143,224,178,228,189,229,132,213,182,222,218,67,229,186,201,169,211,88,210,48,219,152,225,207,62,200,210,220,68,216,44,137,155,181,221,83,214,103,222,183,90,215,86,179,205,128,219,179,164,202,130,224,148,221,146,175,184,213,117,190,201,142,216,82,231,230,92,172,218,207,189,223,150,226,46,216,209,221,216,111,168,219,92,197,228,204,204,101,223,65,211,220,203,122,206,108,224,74,181,142,216,189,190,180,212,107,223,120,191,232,56,209,214,66,209,222,116,172,222,101,179,177,185,168,193,163,194,209,193,105,170,211,212,105,191,117,90,226,90,190,121,216,60,207,67,161,171,220,219,89,184,221,144,182,194,208,214,214,226,196,133,188,202,204,68,194,216,214,77,208,92,215,63,204,178,197,77,115,190,152,227,181,207,102,199,131,208,216,177,116,202,220,121,223,128,196,175,211,219,207,74,218,215,221,93,177,214,183,192,143,139,196,217,226,83,102,119,211,100,212,172,183,180,204,219,226,96,194,47,149,186,143,160,221,224,61,226,147,214,60,198,207,164,146,200,140,182,207,196,208,180,186,215,111,181,221,214,161,216,169,208,136,168,178,112,189,196,124,234,119,202,67,201,74,163,172,148,179,211,143,124,191,202,192,175,218,80,209,72,190,156,217,117,223,122,180,193,182,186,210,219,201,202,224,64,138,192,114,199,208,190,109,218,149,208,97,213,189,87,232,131,152,188,205,219,196,225,94,197,164,210,173,191,222,185,172,189,205,163,208,73,196,201,73,224,206,110,191,122,221,136,165,221,185,156,210,69,210,211,210,102,213,86,170,209,221,197,207,50,154,74,96,212,180,231,158,208,139,150,166,127,213,216,202,95,201,155,217,54,206,214,208,179,140,191,136,228,191,91,232,45,145,220,162,220,74,209,147,155,160,182,217,209,160,174,180,227,219,221,182,100,212,67,213,148,208,51,188,209,82,211,210,71,218,82,176,88,174,117,186,210,160,157,212,135,165,201,162,170,136,176,220,189,167,219,49,212,190,220,73,152,150,176,204,221,124,220,209,117,175,213,228,173,195,159,194,218,57,201,214,164,175,229,79,176,116,206,137,203,152,173,206,78,227,209,154,175,190,89,225,113,222,219,214,50,219,66,202,90,188,214,80,195,74,212,60,130,206,186,157,204,122,198,73,194,230,108,196,205,213,83,192,104,207,117,216,171,126,222,99,229,221,79,214,79,217,144,217,197,206,207,110,160,206,172,197,183,207,217,207,113,210,221,71,161,221,164,227,214,142,177,185,180,103,130,198,123,205,74,216,102,219,160,217,75,204,114,192,213,166,188,118,222,227,92,195,219,161,200,221,69,203,143,198,198,217,198,66,212,50,208,116,199,125,210,207,167,225,116,207,97,184,99,220,184,203,184,219,177,167,202,214,55,207,161,197,122,212,226,187,96,216,201,188,135,224,207,139,225,230,220,121,221,107,212,66,170,169,210,199,102,220,94,159,184,207,92,207,231,214,125,227,220,205,58,193,203,215,223,229,78,196,170,185,196,162,234,56,201,123,171,231,196,86,162,199,213,220,68,200,68,205,88,225,135,220,82,182,215,222,79,152,230,62,162,218,184,224,67,206,99,189,124,214,197,73,204,105,221,179,102,218,232,80,214,181,170,204,165,216,207,217,212,195,176,215,106,192,160,221,182,217,57,211,88,198,233,113,171,204,138,193,209,225,59,176,184,134,223,151,193,200,217,100,225,79,180,142,190,123,222,80,232,216,133,216,148,211,110,198,96,187,224,95,208,112,178,227,94,171,96,181,209,170,225,196,206,94,216,87,217,171,191,82,218,127,227,176,219,207,230,79,214,203,105,213,143,174,188,125,193,220,60,215,172,214,101,211,110,161,117,187,180,125,218,220,62,208,203,217,87,198,156,216,226,161,161,223,224,72,178,198,213,195,219,208,140,175,217,74,201,201,66,186,154,229,89,226,169,204,87,184,85,161,133,201,80,176,188,114,224,77,207,126,202,83,219,200,125,172,169,190,216,80,88,221,68,218,133,216,117,217,157,217,170,190,124,214,210,156,231,84,207,204,113,200,70,222,162,208,227,92,223,136,167,195,221,221,77,173,213,109,214,117,211,217,89,217,91,210,152,194,206,202,110,216,177,190,207,227,185,172,230,172,207,171,199,234,207,149,194,192,179,212,209,210,101,198,225,85,164,211,110,194,182,211,224,65,228,218,79,224,81,122,208,154,129,206,92,193,171,148,188,221,80,220,161,165,166,161,214,99,210,64,174,224,221,105,200,122,230,216,94,223,128,225,161,219,126,187,137,191,222,214,148,151,198,218,210,110,208,228,184,211,35,202,218,195,216,115,212,95,177,199,101,184,208,202,212,134,193,129,192,81,182,223,70,226,230,134,167,183,198,222,227,227,226,63,213,109,187,177,219,223,203,144,179,209,103,177,181,158,221,90,222,166,207,175,230,207,99,205,234,210,210,168,223,143,210,187,209,204,150,209,213,208,193,221,214,77,215,199,81,197,82,177,190,210,231,79,179,221,64,182,199,82,204,204,95,172,187,178,209,86,222,220,118,192,223,88,220,77,174,104,224,137,182,186,96,207,198,74,152,196,217,206,79,214,208,204,180,94,215,81,177,160,201,164,173,205,76,199,220,228,91,215,155,226,79,133,181,136,182,226,96,221,109,209,223,71,202,95,217,87,202,204,183,210,187,212,81,226,184,224,88,170,214,198,226,142,212,81,209,189,172,192,221,216,123,221,126,204,218,222,76,205,73,225,221,73,204,108,201,88,174,197,136,223,90,189,56,207,147,206,212,73,201,83,204,112,137,227,67,208,137,219,225,65,200,186,99,214,97,215,74,203,65,199,216,108,216,80,206,219,104,226,180,225,199,186,197,226,157,102,177,107,231,156,141,226,70,220,216,223,64,214,66,201,174,170,207,46,202,131,173,218,125,217,157,234,192,159,174,209,95,196,224,59,220,69,211,130,203,222,88,208,86,198,127,219,228,75,218,170,168,198,128,215,54,211,167,186,117,211,162,221,219,105,223,99,223,127,202,218,213,143,194,181,200,180,230,224,97,181,132,173,202,221,57,151,220,77,220,160,206,188,101,197,72,213,95,193,212,189,105,226,100,205,201,56,211,93,178,212,88,208,83,213,165,219,183,236,121,220,210,94,212,171,186,218,137,212,129,175,203,223,134,194,95,193,191,105,229,208,102,196,120,191,221,217,65,206,200,74,168,180,199,217,119,223,68,211,125,204,105,180,164,215,227,128,211,166,218,86,185,74,214,57,200,171,111,185,73,199,220,213,192,216,107,211,115,219,227,192,221,101,203,65,211,51,216,84,193,121,214,86,195,115,179,229,90,215,92,207,63,179,212,38,202,104,182,125,179,99,147,184,210,166,227,232,164,120,218,169,203,154,192,224,217,122,160,205,206,221,80,191,217,166,202,78,206,147,202,155,195,76,204,136,191,112,195,160,147,226,91,224,216,212,177,188,165,174,130,203,221,220,133,209,147,216,69,159,155,143,213,94,227,139,209,163,183,199,112,217,213,98,217,96,185,158,173,229,51,209,195,227,214,161,213,83,168,229,209,118,221,224,59,179,161,220,209,193,199,199,212,107,226,219,204,117,166,223,122,166,181,163,176,223,176,130,223,221,202,89,188,147,160,143,218,223,206,151,201,161,130,176,175,138,126,209,112,230,94,211,17,103,218,73,218,131,210,104,214,63,222,38,135,140,215,143,215,191,185,223,207,215,203,46,219,207,93,177,85,213,191,223,56,181,209,82,210,221,66,210,195,223,184,138,217,48,194,73,150,199,220,183,209,60,194,103,218,103,211,216,124,197,217,185,106,185,207,174,165,204,138,220,68,218,151,202,68,214,155,183,221,66,216,61,218,122,214,178,202,178,217,142,215,126,187,148,219,98,180,222,217,80,210,203,43,208,154,220,101,167,206,211,212,208,72,147,225,139,174,207,36,200,234,205,211,180,205,202,126,159,186,116,211,154,192,155,194,168,198,160,218,220,202,153,222,215,66,174,128,211,104,136,171,235,219,112,156,209,109,203,132,192,181,215,112,205,68,215,82,213,117,189,221,186,211,171,208,136,189,128,210,96,199,107,195,232,74,223,132,193,198,46,220,73,181,112,224,133,221,144,224,83,232,217,131,186,53,214,225,95,203,70,102,217,106,224,79,210,113,177,150,228,220,102,225,80,221,170,206,105,223,112,210,46,201,89,197,207,128,235,111,212,161,144,221,182,200,77,213,229,90,134,223,179,212,204,125,197,215,80,233,218,44,226,53,152,184,220,113,219,216,110,214,206,151,215,224,216,163,144,190,133,223,195,216,203,67,95,169,191,131,208,78,104,176,179,148,207,172,220,98,202,118,218,204,120,213,92,213,93,210,203,219,75,212,227,212,188,187,201,100,206,151,200,96,197,215,157,210,70,207,182,205,205,101,212,117,230,86,163,143,167,189,215,168,216,194,98,218,128,219,94,149,188,217,48,172,174,131,131,182,171,200,115,220,217,91,200,80,178,188,226,49,192,205,222,127,194,134,175,214,115,212,214,82,193,106,217,48,197,114,204,114,201,221,190,174,214,168,223,86,217,212,214,181,204,96,190,189,216,91,204,118,226,110,198,224,158,195,117,189,200,150,231,206,91,209,207,118,223,183,232,146,158,207,213,122,204,118,200,103,200,227,76,179,195,73,215,93,214,170,215,232,41,210,107,138,202,204,125,198,134,225,80,117,164,185,197,106,232,74,139,216,207,209,57,207,94,200,229,190,192,140,112,208,155,191,177,216,203,142,192,103,195,219,91,179,228,187,115,213,217,192,193,215,141,218,70,186,37,225,190,84,178,177,162,218,210,185,176,195,97,218,63,218,176,227,215,149,224,221,115,208,214,133,182,188,205,163,207,58,199,217,129,208,184,214,206,133,228,200,80,224,179,229,152,208,95,194,170,224,178,196,181,94,199,90,203,75,216,153,169,213,86,225,67,192,194,65,189,201,103,188,153,163,213,226,142,185,133,226,106,181,215,199,209,211,56,204,114,181,163,171,228,228,73,74,198,203,186,178,185,125,229,221,204,41,165,189,126,205,173,116,179,198,159,216,129,209,222,174,183,229,95,212,68,177,152,217,138,156,135,105,204,193,188,127,209,150,133,163,208,80,218,232,211,77,230,37,198,224,165,213,81,220,207,195,173,174,212,102,206,117,196,178,222,134,205,216,203,86,151,184,157,217,222,123,213,159,195,121,207,159,220,120,212,176,144,217,67,173,216,105,206,90,220,121,217,59,184,219,156,213,149,143,216,221,140,225,182,115,209,115,209,45,223,172,227,133,187,165,199,168,224,91,206,115,153,202,197,62,169,210,134,215,167,203,214,141,200,213,182,90,214,170,206,199,219,54,167,154,72,194,122,181,197,129,214,105,153,209,137,202,227,72,207,61,178,127,181,210,200,46,188,210,214,67,189,216,51,209,125,190,127,208,110,191,219,137,213,76,206,120,186,121,201,222,113,195,194,68,183,179,184,223,61,180,220,197,133,208,226,136,217,200,93,178,220,113,197,198,172,141,225,102,159,149,213,196,100,220,196,176,232,182,187,171,165,182,101,175,169,191,224,110,200,128,200,129,114,179,188,165,198,216,184,174,216,67,229,198,220,32,232,219,72,219,203,127,88,212,81,142,223,210,166,97,145,209,77,216,227,196,83,202,137,214,82,223,114,205,177,183,196,214,129,196,122,223,157,232,99,180,188,203,132,229,223,186,115,209,191,218,50,192,184,220,102,207,87,196,162,219,92,221,140,217,139,169,213,79,211,99,205,104,200,86,210,90,157,151,227,228,53,205,72,195,75,226,89,226,74,218,145,228,224,208,171,215,153,140,208,182,161,228,107,209,220,217,207,125,181,195,212,220,95,202,95,191,233,74,201,184,221,81,231,181,120,227,119,139,121,179,199,203,216,154,210,144,195,129,153,213,103,209,219,212,125,216,229,219,108,223,65,212,92,221,197,162,211,147,210,197,178,221,162,192,172,215,84,194,52,204,70,175,187,187,194,186,235,185,177,170,216,213,64,212,102,191,112,143,204,96,164,226,218,107,182,116,224,157,223,171,194,104,228,114,218,40,207,54,204,220,108,199,214,195,81,158,130,133,116,118,203,215,215,146,219,210,69,216,121,225,59,210,65,217,202,79,209,76,156,152,178,193,86,228,50,217,94,220,71,156,206,84,202,88,113,211,215,65,168,221,195,219,213,148,204,84,212,217,184,218,228,118,222,76,222,226,205,126,218,224,216,77,122,218,74,213,81,220,66,190,132,212,213,200,216,63,204,105,169,166,173,193,142,201,168,213,96,200,210,93,210,217,214,182,204,199,172,159,200,216,162,198,168,191,200,99,214,227,230,87,171,204,195,176,210,165,158,221,118,195,182,217,225,217,191,210,120,164,100,222,195,227,81,229,174,207,212,92,191,204,206,61,209,143,230,113,218,224,188,223,95,216,71,154,212,107,150,171,192,204,66,216,218,113,218,206,155,187,185,209,180,111,230,66,206,89,225,62,218,183,201,229,215,172,196,114,199,211,54,203,198,211,110,223,85,219,133,133,221,154,155,186,205,77,36,213,221,196,174,180,135,217,54,181,219,215,194,206,154,216,101,209,162,184,192,147,224,84,183,223,68,197,155,205,174,97,199,177,211,221,68,204,103,221,163,203,109,186,216,216,152,228,67,218,100,140,222,113,202,70,202,183,125,126,203,226,175,165,160,182,121,172,231,204,185,159,198,219,227,150,194,108,204,72,215,119,212,190,223,213,63,234,116,234,180,218,182,189,175,205,110,226,213,184,191,88,182,162,187,163,193,198,49,214,148,159,214,162,154,227,187,123,185,167,184,127,229,228,147,150,187,222,104,217,153,229,211,112,179,198,211,124,222,166,207,99,152,175,160,214,176,134,213,74,201,93,171,202,183,212,91,208,60,221,164,207,72,181,167,210,170,226,50,147,206,80,209,136,210,108,180,221,218,105,213,118,188,216,219,94,216,162,219,99,209,207,206,217,221,221,222,56,182,222,220,229,226,209,111,231,144,229,126,210,140,187,155,214,138,226,128,213,193,99,159,115,218,98,219,96]
 
1
+ [206,104,226,67,200,185,221,212,206,222,88,210,228,174,155,205,172,218,148,132,212,91,163,184,205,132,213,190,212,198,230,227,159,198,122,216,175,197,118,217,219,224,69,220,197,72,204,92,169,191,191,155,175,111,218,77,207,36,195,178,123,170,231,91,209,177,146,205,151,221,217,95,199,89,153,216,154,202,167,213,104,184,176,226,200,232,53,223,73,202,220,158,174,189,165,222,158,221,211,78,205,72,214,212,215,232,175,219,110,205,192,226,34,176,209,158,222,132,179,208,98,126,205,126,172,167,227,100,224,182,208,117,217,199,195,191,169,178,158,217,152,157,163,163,204,209,146,217,150,194,217,23,125,200,221,200,212,41,176,223,207,95,178,232,68,182,173,205,198,210,139,152,147,215,196,223,83,199,123,197,119,230,223,65,216,85,210,52,210,204,179,111,138,215,83,177,219,69,212,77,182,226,99,178,207,197,87,190,222,216,85,208,211,66,220,226,221,134,175,190,170,166,157,216,135,211,132,200,67,227,195,178,221,179,205,168,186,208,127,207,139,205,67,209,223,117,195,190,202,64,218,77,199,221,189,125,216,80,148,214,143,181,98,221,194,112,219,212,138,216,87,127,187,219,122,220,97,223,221,164,205,220,100,227,101,171,188,223,89,213,137,175,172,219,77,102,234,109,197,114,205,184,221,138,175,152,163,133,227,73,224,181,174,147,144,178,147,224,211,121,123,211,219,209,210,204,80,182,215,152,118,224,216,154,220,163,204,187,133,151,162,221,94,184,224,213,72,187,148,223,195,110,182,117,224,88,202,162,220,154,151,197,227,197,74,221,205,97,221,219,103,206,184,72,191,201,192,230,164,217,170,214,118,221,218,118,215,181,224,78,183,154,190,206,102,202,74,195,96,225,228,221,99,224,104,123,203,184,211,90,209,173,199,203,99,201,176,160,207,132,231,96,195,213,173,101,190,206,61,225,232,215,178,211,227,192,148,195,213,114,212,127,162,214,213,129,220,70,176,224,209,137,232,116,228,212,75,182,207,71,111,173,189,207,96,172,213,119,222,93,195,204,211,76,179,200,184,216,107,135,202,71,118,219,123,202,210,146,225,76,227,62,214,226,98,158,202,219,169,83,156,223,125,211,213,102,219,223,72,224,178,190,204,110,228,64,214,200,193,205,213,160,201,216,134,195,128,186,213,219,187,199,216,84,202,150,211,127,230,206,204,180,183,183,167,213,224,72,209,79,195,205,153,228,133,203,210,78,190,207,195,133,225,192,197,190,201,187,222,109,138,227,103,223,209,164,190,73,226,200,205,191,189,100,226,135,198,129,185,207,229,148,209,216,58,222,117,219,92,221,147,222,219,63,216,228,165,181,157,213,197,218,158,191,186,223,204,109,172,229,91,231,204,102,192,217,204,144,222,173,209,209,95,186,195,60,212,219,209,162,183,227,216,111,193,182,222,71,183,222,122,178,219,130,228,82,225,223,69,198,232,63,193,207,216,139,223,74,164,198,224,80,148,227,138,227,199,225,86,148,230,169,210,92,166,187,202,165,196,78,202,74,229,173,216,215,213,164,205,102,206,91,212,173,198,109,183,94,206,115,224,126,229,222,57,225,191,221,218,215,44,177,217,214,218,191,221,80,198,109,196,95,192,181,214,214,123,208,191,225,219,218,108,204,150,199,206,90,227,71,207,123,199,106,131,213,93,178,208,60,132,217,149,204,80,227,218,202,94,232,175,213,115,217,114,222,200,173,184,217,222,178,226,205,192,185,197,180,193,205,211,136,221,62,184,213,89,225,71,169,216,135,198,98,214,204,88,191,154,207,213,227,93,230,133,216,189,122,224,108,229,237,218,209,214,66,153,203,219,204,222,218,107,174,214,148,169,188,210,98,203,142,167,174,224,103,223,220,148,216,199,188,175,173,130,156,97,220,111,216,115,216,70,221,210,77,173,162,216,164,216,69,203,175,172,166,199,156,184,131,201,210,98,220,133,211,116,221,82,213,88,213,128,227,219,177,190,216,180,237,217,64,215,134,221,172,220,221,70,169,221,189,226,74,229,197,209,128,190,213,106,221,198,162,184,182,219,86,201,143,229,205,146,228,106,154,209,60,216,222,74,221,215,141,168,202,82,194,213,80,217,214,204,164,199,220,149,114,140,215,150,218,108,109,211,170,206,125,183,148,193,229,210,124,188,143,160,153,199,178,75,225,105,199,83,222,112,205,75,222,205,70,104,194,91,158,220,99,215,209,219,65,124,215,158,223,234,208,101,218,200,66,188,222,149,201,189,211,51,179,208,198,160,182,152,195,141,216,112,205,207,170,128,219,118,188,223,224,80,177,218,127,196,151,159,117,186,113,216,212,134,204,90,200,231,184,222,68,146,151,158,202,173,144,206,171,215,68,188,222,89,215,117,223,80,191,209,204,105,228,67,174,210,74,208,105,151,159,183,142,163,202,214,41,147,222,227,188,208,200,157,132,195,94,191,185,214,132,201,155,166,195,220,147,219,156,214,165,220,105,223,217,100,176,225,117,207,134,166,202,218,172,200,157,183,217,87,214,181,229,217,76,231,199,139,154,213,208,208,169,218,82,219,191,139,207,222,49,122,154,176,219,110,185,112,137,231,80,208,136,189,192,208,190,192,187,180,220,186,225,218,208,102,190,124,152,83,219,199,95,198,133,235,214,96,213,217,148,168,200,32,145,207,222,198,171,208,218,104,178,230,77,214,129,218,136,139,212,233,93,221,153,221,95,155,151,211,225,60,81,217,201,108,208,234,172,177,172,193,208,72,216,224,211,44,205,82,221,224,199,198,221,137,208,81,183,105,182,106,222,180,184,146,180,205,187,192,163,219,213,164,229,218,190,147,221,193,213,182,126,228,231,97,173,218,116,161,156,207,122,193,198,155,207,103,223,192,92,153,223,170,206,177,225,40,224,106,203,208,116,203,105,197,152,221,211,84,226,111,215,154,224,221,65,215,142,169,212,184,135,207,85,209,213,76,225,229,130,203,217,186,229,140,177,116,216,219,62,209,105,164,150,207,214,211,114,148,230,118,201,134,197,124,209,163,184,164,186,189,219,106,221,220,95,141,217,87,200,224,114,220,168,212,112,229,192,106,189,183,209,212,85,187,202,219,222,223,223,212,226,79,193,128,222,209,206,197,178,152,114,222,111,164,183,189,179,214,130,149,150,212,122,160,195,118,115,204,144,163,159,140,211,208,120,222,220,125,195,174,220,80,173,204,87,149,212,82,221,119,213,139,186,227,97,215,213,136,215,68,229,218,63,224,182,221,218,94,213,90,168,202,85,168,222,121,215,213,93,221,83,220,112,197,213,213,92,210,76,216,79,179,220,94,194,99,186,181,197,214,194,138,221,91,218,172,159,219,74,153,221,201,108,213,131,217,112,214,107,209,35,224,188,219,108,191,105,157,211,126,206,148,220,126,217,164,203,129,201,211,119,204,165,209,75,205,130,204,215,206,119,178,211,218,137,193,167,199,220,182,115,207,184,189,217,178,186,159,208,230,140,194,75,173,205,64,213,167,189,110,220,195,136,137,196,134,208,120,198,182,189,176,162,183,172,172,214,200,95,212,189,193,184,195,116,167,199,110,180,217,126,193,226,142,204,66,140,222,226,218,144,218,214,140,206,66,196,224,143,121,166,221,101,154,204,216,201,221,152,225,69,226,226,180,200,181,212,128,193,206,136,219,141,208,128,202,209,105,168,178,204,211,110,145,163,193,211,69,220,147,194,223,165,212,218,124,197,229,226,132,220,156,211,80,203,140,192,94,208,102,180,215,134,152,218,106,191,204,102,211,108,182,204,206,208,90,165,197,105,186,115,213,208,186,165,186,174,213,211,164,210,62,129,201,217,170,169,217,193,167,222,174,217,167,128,182,113,159,208,85,172,219,89,219,126,218,190,65,227,208,52,225,125,160,205,135,168,186,213,212,69,124,205,67,204,110,151,180,220,205,193,106,223,80,124,166,218,194,138,148,205,88,179,183,163,227,221,198,213,87,190,200,239,81,202,211,213,115,204,105,216,83,170,201,210,95,219,219,180,216,222,228,86,231,83,227,116,223,84,207,216,115,204,109,194,171,198,183,197,86,216,144,209,125,226,221,57,215,117,199,185,152,230,174,220,61,209,99,163,218,218,134,183,212,85,208,92,190,214,81,146,220,181,224,214,212,79,171,209,217,209,100,213,72,220,223,193,72,130,203,111,214,192,197,222,201,229,204,189,162,124,221,157,218,34,204,88,170,205,76,168,210,46,196,223,73,180,85,118,200,85,221,68,160,202,201,219,181,148,207,210,208,69,210,101,215,76,217,104,188,180,224,71,199,166,203,106,216,196,182,84,183,125,128,224,89,206,88,212,212,152,205,92,210,65,166,171,200,120,228,213,83,163,131,153,180,215,228,99,179,210,82,218,95,209,94,215,206,196,110,236,219,209,193,95,172,177,218,218,140,203,228,102,160,215,102,204,192,89,168,222,56,223,155,189,213,208,220,78,204,213,168,202,169,230,191,218,203,73,222,127,208,143,221,217,217,100,160,212,208,156,223,197,205,201,223,220,144,209,90,209,139,182,218,22,221,183,219,79,216,194,224,86,208,225,190,102,200,212,162,211,83,221,106,215,113,161,197,111,216,210,92,219,124,208,185,152,203,220,230,106,182,193,185,221,79,197,79,231,192,202,113,166,211,142,195,170,179,226,222,208,228,152,133,200,89,197,180,155,215,97,201,139,205,145,217,163,213,85,223,209,200,139,217,123,160,183,201,136,210,206,105,208,132,206,208,101,134,174,112,205,181,221,215,221,213,225,116,217,80,162,186,120,183,122,214,201,97,188,209,174,229,205,120,194,203,130,228,134,228,83,201,119,218,210,55,225,64,171,174,227,107,226,95,139,190,154,171,206,139,154,219,157,224,213,206,212,79,135,186,156,217,186,128,118,184,187,213,221,86,183,217,218,210,216,155,195,188,134,172,223,125,176,167,85,187,214,217,94,213,128,217,133,192,217,60,189,122,217,117,224,95,199,74,206,89,221,205,158,196,77,228,114,188,105,210,70,209,83,188,169,214,225,184,205,127,197,115,182,174,181,238,222,223,59,229,219,112,205,115,190,128,196,210,222,176,215,194,187,112,212,90,183,207,86,212,85,178,187,138,217,80,203,96,221,198,82,207,96,164,227,208,76,214,179,209,186,207,205,221,224,143,224,178,228,189,229,132,213,182,222,218,67,229,186,201,169,211,88,210,48,219,152,225,207,62,200,210,220,68,216,44,137,155,181,221,83,214,103,222,183,90,215,86,179,205,128,219,179,164,202,130,224,148,221,146,175,184,213,117,190,201,142,216,82,231,230,92,172,218,207,189,223,150,226,46,216,209,221,216,111,168,219,92,197,228,204,204,101,223,65,211,220,203,122,206,108,224,74,181,142,216,189,190,180,212,107,223,120,191,232,56,209,214,66,209,222,116,172,222,101,179,177,185,168,193,163,194,209,193,105,170,211,212,105,191,117,90,226,90,190,121,216,60,207,67,161,171,220,219,89,184,221,144,182,194,208,214,214,226,196,133,188,202,204,68,194,216,214,77,208,92,215,63,204,178,197,77,115,190,152,227,181,207,102,199,131,208,216,177,116,202,220,121,223,128,196,175,211,219,207,74,218,215,221,93,177,214,183,192,143,139,196,217,226,83,102,119,211,100,212,172,183,180,204,219,226,96,194,47,149,186,143,160,221,224,61,226,147,214,60,198,207,164,146,200,140,182,207,196,208,180,186,215,111,181,221,214,161,216,169,208,136,168,178,112,189,196,124,234,119,202,67,201,74,163,172,148,179,211,143,124,191,202,192,175,218,80,209,72,190,156,217,117,223,122,180,193,182,186,210,219,201,202,224,64,138,192,114,199,208,190,109,218,149,208,97,213,189,87,232,131,152,188,205,219,196,225,94,197,164,210,173,191,222,185,172,189,205,163,208,73,196,201,73,224,206,110,191,122,221,136,165,221,185,156,210,69,210,211,210,102,213,86,170,209,221,197,207,50,154,74,96,212,180,231,158,208,139,150,166,127,213,216,202,95,201,155,217,54,206,214,208,179,140,191,136,228,191,91,232,45,145,220,162,220,74,209,147,155,160,182,217,209,160,174,180,227,219,221,182,100,212,67,213,148,208,51,188,209,82,211,210,71,218,82,176,88,174,117,186,210,160,157,212,135,165,201,162,170,136,176,220,189,167,219,49,212,190,220,73,152,150,176,204,221,124,220,209,117,175,213,228,173,195,159,194,218,57,201,214,164,175,229,79,176,116,206,137,203,152,173,206,78,227,209,154,175,190,89,225,113,222,219,214,50,219,66,202,90,188,214,80,195,74,212,60,130,206,186,157,204,122,198,73,194,230,108,196,205,213,83,192,104,207,117,216,171,126,222,99,229,221,79,214,79,217,144,217,197,206,207,110,160,206,172,197,183,207,217,207,113,210,221,71,161,221,164,227,214,142,177,185,180,103,130,198,123,205,74,216,102,219,160,217,75,204,114,192,213,166,188,118,222,227,92,195,219,161,200,221,69,203,143,198,198,217,198,66,212,50,208,116,199,125,210,207,167,225,116,207,97,184,99,220,184,203,184,219,177,167,202,214,55,207,161,197,122,212,226,187,96,216,201,188,135,224,207,139,225,230,220,121,221,107,212,66,170,169,210,199,102,220,94,159,184,207,92,207,231,214,125,227,220,205,58,193,203,215,223,229,78,196,170,185,196,162,234,56,201,123,171,231,196,86,162,199,213,220,68,200,68,205,88,225,135,220,82,182,215,222,79,152,230,62,162,218,184,224,67,206,99,189,124,214,197,73,204,105,221,179,102,218,232,80,214,181,170,204,165,216,207,217,212,195,176,215,106,192,160,221,182,217,57,211,88,198,233,113,171,204,138,193,209,225,59,176,184,134,223,151,193,200,217,100,225,79,180,142,190,123,222,80,232,216,133,216,148,211,110,198,96,187,224,95,208,112,178,227,94,171,96,181,209,170,225,196,206,94,216,87,217,171,191,82,218,127,227,176,219,207,230,79,214,203,105,213,143,174,188,125,193,220,60,215,172,214,101,211,110,161,117,187,180,125,218,220,62,208,203,217,87,198,156,216,226,161,161,223,224,72,178,198,213,195,219,208,140,175,217,74,201,201,66,186,154,229,89,226,169,204,87,184,85,161,133,201,80,176,188,114,224,77,207,126,202,83,219,200,125,172,169,190,216,80,88,221,68,218,133,216,117,217,157,217,170,190,124,214,210,156,231,84,207,204,113,200,70,222,162,208,227,92,223,136,167,195,221,221,77,173,213,109,214,117,211,217,89,217,91,210,152,194,206,202,110,216,177,190,207,227,185,172,230,172,207,171,199,234,207,149,194,192,179,212,209,210,101,198,225,85,164,211,110,194,182,211,224,65,228,218,79,224,81,122,208,154,129,206,92,193,171,148,188,221,80,220,161,165,166,161,214,99,210,64,174,224,221,105,200,122,230,216,94,223,128,225,161,219,126,187,137,191,222,214,148,151,198,218,210,110,208,228,184,211,35,202,218,195,216,115,212,95,177,199,101,184,208,202,212,134,193,129,192,81,182,223,70,226,230,134,167,183,198,222,227,227,226,63,213,109,187,177,219,223,203,144,179,209,103,177,181,158,221,90,222,166,207,175,230,207,99,205,234,210,210,168,223,143,210,187,209,204,150,209,213,208,193,221,214,77,215,199,81,197,82,177,190,210,231,79,179,221,64,182,199,82,204,204,95,172,187,178,209,86,222,220,118,192,223,88,220,77,174,104,224,137,182,186,96,207,198,74,152,196,217,206,79,214,208,204,180,94,215,81,177,160,201,164,173,205,76,199,220,228,91,215,155,226,79,133,181,136,182,226,96,221,109,209,223,71,202,95,217,87,202,204,183,210,187,212,81,226,184,224,88,170,214,198,226,142,212,81,209,189,172,192,221,216,123,221,126,204,218,222,76,205,73,225,221,73,204,108,201,88,174,197,136,223,90,189,56,207,147,206,212,73,201,83,204,112,137,227,67,208,137,219,225,65,200,186,99,214,97,215,74,203,65,199,216,108,216,80,206,219,104,226,180,225,199,186,197,226,157,102,177,107,231,156,141,226,70,220,216,223,64,214,66,201,174,170,207,46,202,131,173,218,125,217,157,234,192,159,174,209,95,196,224,59,220,69,211,130,203,222,88,208,86,198,127,219,228,75,218,170,168,198,128,215,54,211,167,186,117,211,162,221,219,105,223,99,223,127,202,218,213,143,194,181,200,180,230,224,97,181,132,173,202,221,57,151,220,77,220,160,206,188,101,197,72,213,95,193,212,189,105,226,100,205,201,56,211,93,178,212,88,208,83,213,165,219,183,236,121,220,210,94,212,171,186,218,137,212,129,175,203,223,134,194,95,193,191,105,229,208,102,196,120,191,221,217,65,206,200,74,168,180,199,217,119,223,68,211,125,204,105,180,164,215,227,128,211,166,218,86,185,74,214,57,200,171,111,185,73,199,220,213,192,216,107,211,115,219,227,192,221,101,203,65,211,51,216,84,193,121,214,86,195,115,179,229,90,215,92,207,63,179,212,38,202,104,182,125,179,99,147,184,210,166,227,232,164,120,218,169,203,154,192,224,217,122,160,205,206,221,80,191,217,166,202,78,206,147,202,155,195,76,204,136,191,112,195,160,147,226,91,224,216,212,177,188,165,174,130,203,221,220,133,209,147,216,69,159,155,143,213,94,227,139,209,163,183,199,112,217,213,98,217,96,185,158,173,229,51,209,195,227,214,161,213,83,168,229,209,118,221,224,59,179,161,220,209,193,199,199,212,107,226,219,204,117,166,223,122,166,181,163,176,223,176,130,223,221,202,89,188,147,160,143,218,223,206,151,201,161,130,176,175,138,126,209,112,230,94,211,17,103,218,73,218,131,210,104,214,63,222,38,135,140,215,143,215,191,185,223,207,215,203,46,219,207,93,177,85,213,191,223,56,181,209,82,210,221,66,210,195,223,184,138,217,48,194,73,150,199,220,183,209,60,194,103,218,103,211,216,124,197,217,185,106,185,207,174,165,204,138,220,68,218,151,202,68,214,155,183,221,66,216,61,218,122,214,178,202,178,217,142,215,126,187,148,219,98,180,222,217,80,210,203,43,208,154,220,101,167,206,211,212,208,72,147,225,139,174,207,36,200,234,205,211,180,205,202,126,159,186,116,211,154,192,155,194,168,198,160,218,220,202,153,222,215,66,174,128,211,104,136,171,235,219,112,156,209,109,203,132,192,181,215,112,205,68,215,82,213,117,189,221,186,211,171,208,136,189,128,210,96,199,107,195,232,74,223,132,193,198,46,220,73,181,112,224,133,221,144,224,83,232,217,131,186,53,214,225,95,203,70,102,217,106,224,79,210,113,177,150,228,220,102,225,80,221,170,206,105,223,112,210,46,201,89,197,207,128,235,111,212,161,144,221,182,200,77,213,229,90,134,223,179,212,204,125,197,215,80,233,218,44,226,53,152,184,220,113,219,216,110,214,206,151,215,224,216,163,144,190,133,223,195,216,203,67,95,169,191,131,208,78,104,176,179,148,207,172,220,98,202,118,218,204,120,213,92,213,93,210,203,219,75,212,227,212,188,187,201,100,206,151,200,96,197,215,157,210,70,207,182,205,205,101,212,117,230,86,163,143,167,189,215,168,216,194,98,218,128,219,94,149,188,217,48,172,174,131,131,182,171,200,115,220,217,91,200,80,178,188,226,49,192,205,222,127,194,134,175,214,115,212,214,82,193,106,217,48,197,114,204,114,201,221,190,174,214,168,223,86,217,212,214,181,204,96,190,189,216,91,204,118,226,110,198,224,158,195,117,189,200,150,231,206,91,209,207,118,223,183,232,146,158,207,213,122,204,118,200,103,200,227,76,179,195,73,215,93,214,170,215,232,41,210,107,138,202,204,125,198,134,225,80,117,164,185,197,106,232,74,139,216,207,209,57,207,94,200,229,190,192,140,112,208,155,191,177,216,203,142,192,103,195,219,91,179,228,187,115,213,217,192,193,215,141,218,70,186,37,225,190,84,178,177,162,218,210,185,176,195,97,218,63,218,176,227,215,149,224,221,115,208,214,133,182,188,205,163,207,58,199,217,129,208,184,214,206,133,228,200,80,224,179,229,152,208,95,194,170,224,178,196,181,94,199,90,203,75,216,153,169,213,86,225,67,192,194,65,189,201,103,188,153,163,213,226,142,185,133,226,106,181,215,199,209,211,56,204,114,181,163,171,228,228,73,74,198,203,186,178,185,125,229,221,204,41,165,189,126,205,173,116,179,198,159,216,129,209,222,174,183,229,95,212,68,177,152,217,138,156,135,105,204,193,188,127,209,150,133,163,208,80,218,232,211,77,230,37,198,224,165,213,81,220,207,195,173,174,212,102,206,117,196,178,222,134,205,216,203,86,151,184,157,217,222,123,213,159,195,121,207,159,220,120,212,176,144,217,67,173,216,105,206,90,220,121,217,59,184,219,156,213,149,143,216,221,140,225,182,115,209,115,209,45,223,172,227,133,187,165,199,168,224,91,206,115,153,202,197,62,169,210,134,215,167,203,214,141,200,213,182,90,214,170,206,199,219,54,167,154,72,194,122,181,197,129,214,105,153,209,137,202,227,72,207,61,178,127,181,210,200,46,188,210,214,67,189,216,51,209,125,190,127,208,110,191,219,137,213,76,206,120,186,121,201,222,113,195,194,68,183,179,184,223,61,180,220,197,133,208,226,136,217,200,93,178,220,113,197,198,172,141,225,102,159,149,213,196,100,220,196,176,232,182,187,171,165,182,101,175,169,191,224,110,200,128,200,129,114,179,188,165,198,216,184,174,216,67,229,198,220,32,232,219,72,219,203,127,88,212,81,142,223,210,166,97,145,209,77,216,227,196,83,202,137,214,82,223,114,205,177,183,196,214,129,196,122,223,157,232,99,180,188,203,132,229,223,186,115,209,191,218,50,192,184,220,102,207,87,196,162,219,92,221,140,217,139,169,213,79,211,99,205,104,200,86,210,90,157,151,227,228,53,205,72,195,75,226,89,226,74,218,145,228,224,208,171,215,153,140,208,182,161,228,107,209,220,217,207,125,181,195,212,220,95,202,95,191,233,74,201,184,221,81,231,181,120,227,119,139,121,179,199,203,216,154,210,144,195,129,153,213,103,209,219,212,125,216,229,219,108,223,65,212,92,221,197,162,211,147,210,197,178,221,162,192,172,215,84,194,52,204,70,175,187,187,194,186,235,185,177,170,216,213,64,212,102,191,112,143,204,96,164,226,218,107,182,116,224,157,223,171,194,104,228,114,218,40,207,54,204,220,108,199,214,195,81,158,130,133,116,118,203,215,215,146,219,210,69,216,121,225,59,210,65,217,202,79,209,76,156,152,178,193,86,228,50,217,94,220,71,156,206,84,202,88,113,211,215,65,168,221,195,219,213,148,204,84,212,217,184,218,228,118,222,76,222,226,205,126,218,224,216,77,122,218,74,213,81,220,66,190,132,212,213,200,216,63,204,105,169,166,173,193,142,201,168,213,96,200,210,93,210,217,214,182,204,199,172,159,200,216,162,198,168,191,200,99,214,227,230,87,171,204,195,176,210,165,158,221,118,195,182,217,225,217,191,210,120,164,100,222,195,227,81,229,174,207,212,92,191,204,206,61,209,143,230,113,218,224,188,223,95,216,71,154,212,107,150,171,192,204,66,216,218,113,218,206,155,187,185,209,180,111,230,66,206,89,225,62,218,183,201,229,215,172,196,114,199,211,54,203,198,211,110,223,85,219,133,133,221,154,155,186,205,77,36,213,221,196,174,180,135,217,54,181,219,215,194,206,154,216,101,209,162,184,192,147,224,84,183,223,68,197,155,205,174,97,199,177,211,221,68,204,103,221,163,203,109,186,216,216,152,228,67,218,100,140,222,113,202,70,202,183,125,126,203,226,175,165,160,182,121,172,231,204,185,159,198,219,227,150,194,108,204,72,215,119,212,190,223,213,63,234,116,234,180,218,182,189,175,205,110,226,213,184,191,88,182,162,187,163,193,198,49,214,148,159,214,162,154,227,187,123,185,167,184,127,229,228,147,150,187,222,104,217,153,229,211,112,179,198,211,124,222,166,207,99,152,175,160,214,176,134,213,74,201,93,171,202,183,212,91,208,60,221,164,207,72,181,167,210,170,226,50,147,206,80,209,136,210,108,180,221,218,105,213,118,188,216,219,94,216,162,219,99,209,207,206,217,221,221,222,56,182,222,220,229,226,209,111,231,144,229,126,210,140,187,155,214,138,226,128,213,193,99,159,115,218,98,219,96,152]
ivf.pid.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:1a92d38f8681636364e2408f1768d9292f361e3cffcfc5db3582ccc18e663433
3
- size 2509016
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:02faa50987b84db3735173b86f73e481d3d9fc6d896fded589fb4e17a295ae45
3
+ size 2508568
metadata.json CHANGED
@@ -37,7 +37,7 @@
37
  "checkpoint":"colbert-ir/colbertv2.0",
38
  "triples":"/future/u/okhattab/root/unit/experiments/2021.10/downstream.distillation.round2.2_score/round2.nway6.cosine.ib/examples.64.json",
39
  "collection":[
40
- "list with 5497 elements starting with...",
41
  [
42
  "Driven by large-data pre-training, Segment Anything Model (SAM) has been demonstrated as a powerful and promptable framework, revolutionizing the segmentation models. Despite the generality, customizing SAM for specific visual concepts without man-powered prompting is under explored, e.g., automatically segmenting your pet dog in different images. In this paper, we propose a training-free Personalization approach for SAM, termed as PerSAM. Given only a single image with a reference mask, PerSAM first localizes the target concept by a location prior, and segments it within other images or videos via three techniques: target-guided attention, target-semantic prompting, and cascaded post-refinement. In this way, we effectively adapt SAM for private use without any training. To further alleviate the mask ambiguity, we present an efficient one-shot fine-tuning variant, PerSAM-F. Freezing the entire SAM, we introduce two learnable weights for multi-scale masks, only training 2 parameters within 10 seconds for improved performance. To demonstrate our efficacy, we construct a new segmentation dataset, PerSeg, for personalized evaluation, and test our methods on video object segmentation with competitive performance.",
43
  "Freezing the entire SAM, we introduce two learnable weights for multi-scale masks, only training 2 parameters within 10 seconds for improved performance. To demonstrate our efficacy, we construct a new segmentation dataset, PerSeg, for personalized evaluation, and test our methods on video object segmentation with competitive performance. Besides, our approach can also enhance DreamBooth to personalize Stable Diffusion for text-to-image generation, which discards the background disturbance for better target appearance learning. Code is released at https://github.com/ZrrSkywalker/Personalize-SAM",
@@ -50,7 +50,7 @@
50
  "root":".ragatouille/",
51
  "experiment":"colbert",
52
  "index_root":null,
53
- "name":"2024-10/09/08.54.10",
54
  "rank":0,
55
  "nranks":1,
56
  "amp":true,
@@ -59,8 +59,8 @@
59
  },
60
  "num_chunks":1,
61
  "num_partitions":8192,
62
- "num_embeddings":941392,
63
- "avg_doclen":171.2555939603,
64
  "RAGatouille":{
65
  "index_config":{
66
  "index_type":"PLAID",
 
37
  "checkpoint":"colbert-ir/colbertv2.0",
38
  "triples":"/future/u/okhattab/root/unit/experiments/2021.10/downstream.distillation.round2.2_score/round2.nway6.cosine.ib/examples.64.json",
39
  "collection":[
40
+ "list with 5498 elements starting with...",
41
  [
42
  "Driven by large-data pre-training, Segment Anything Model (SAM) has been demonstrated as a powerful and promptable framework, revolutionizing the segmentation models. Despite the generality, customizing SAM for specific visual concepts without man-powered prompting is under explored, e.g., automatically segmenting your pet dog in different images. In this paper, we propose a training-free Personalization approach for SAM, termed as PerSAM. Given only a single image with a reference mask, PerSAM first localizes the target concept by a location prior, and segments it within other images or videos via three techniques: target-guided attention, target-semantic prompting, and cascaded post-refinement. In this way, we effectively adapt SAM for private use without any training. To further alleviate the mask ambiguity, we present an efficient one-shot fine-tuning variant, PerSAM-F. Freezing the entire SAM, we introduce two learnable weights for multi-scale masks, only training 2 parameters within 10 seconds for improved performance. To demonstrate our efficacy, we construct a new segmentation dataset, PerSeg, for personalized evaluation, and test our methods on video object segmentation with competitive performance.",
43
  "Freezing the entire SAM, we introduce two learnable weights for multi-scale masks, only training 2 parameters within 10 seconds for improved performance. To demonstrate our efficacy, we construct a new segmentation dataset, PerSeg, for personalized evaluation, and test our methods on video object segmentation with competitive performance. Besides, our approach can also enhance DreamBooth to personalize Stable Diffusion for text-to-image generation, which discards the background disturbance for better target appearance learning. Code is released at https://github.com/ZrrSkywalker/Personalize-SAM",
 
50
  "root":".ragatouille/",
51
  "experiment":"colbert",
52
  "index_root":null,
53
+ "name":"2024-10/09/09.54.21",
54
  "rank":0,
55
  "nranks":1,
56
  "amp":true,
 
59
  },
60
  "num_chunks":1,
61
  "num_partitions":8192,
62
+ "num_embeddings":941544,
63
+ "avg_doclen":171.2520916697,
64
  "RAGatouille":{
65
  "index_config":{
66
  "index_type":"PLAID",
pid_docid_map.json CHANGED
@@ -5495,5 +5495,6 @@
5495
  "5493":"2410.04717",
5496
  "5494":"2410.04717",
5497
  "5495":"2410.02743",
5498
- "5496":"2410.02743"
 
5499
  }
 
5495
  "5493":"2410.04717",
5496
  "5494":"2410.04717",
5497
  "5495":"2410.02743",
5498
+ "5496":"2410.02743",
5499
+ "5497":"2410.04422"
5500
  }
plan.json CHANGED
@@ -37,7 +37,7 @@
37
  "checkpoint": "colbert-ir\/colbertv2.0",
38
  "triples": "\/future\/u\/okhattab\/root\/unit\/experiments\/2021.10\/downstream.distillation.round2.2_score\/round2.nway6.cosine.ib\/examples.64.json",
39
  "collection": [
40
- "list with 5497 elements starting with...",
41
  [
42
  "Driven by large-data pre-training, Segment Anything Model (SAM) has been demonstrated as a powerful and promptable framework, revolutionizing the segmentation models. Despite the generality, customizing SAM for specific visual concepts without man-powered prompting is under explored, e.g., automatically segmenting your pet dog in different images. In this paper, we propose a training-free Personalization approach for SAM, termed as PerSAM. Given only a single image with a reference mask, PerSAM first localizes the target concept by a location prior, and segments it within other images or videos via three techniques: target-guided attention, target-semantic prompting, and cascaded post-refinement. In this way, we effectively adapt SAM for private use without any training. To further alleviate the mask ambiguity, we present an efficient one-shot fine-tuning variant, PerSAM-F. Freezing the entire SAM, we introduce two learnable weights for multi-scale masks, only training 2 parameters within 10 seconds for improved performance. To demonstrate our efficacy, we construct a new segmentation dataset, PerSeg, for personalized evaluation, and test our methods on video object segmentation with competitive performance.",
43
  "Freezing the entire SAM, we introduce two learnable weights for multi-scale masks, only training 2 parameters within 10 seconds for improved performance. To demonstrate our efficacy, we construct a new segmentation dataset, PerSeg, for personalized evaluation, and test our methods on video object segmentation with competitive performance. Besides, our approach can also enhance DreamBooth to personalize Stable Diffusion for text-to-image generation, which discards the background disturbance for better target appearance learning. Code is released at https:\/\/github.com\/ZrrSkywalker\/Personalize-SAM",
@@ -50,7 +50,7 @@
50
  "root": ".ragatouille\/",
51
  "experiment": "colbert",
52
  "index_root": null,
53
- "name": "2024-10\/09\/08.54.10",
54
  "rank": 0,
55
  "nranks": 1,
56
  "amp": true,
@@ -59,6 +59,6 @@
59
  },
60
  "num_chunks": 1,
61
  "num_partitions": 8192,
62
- "num_embeddings_est": 941392.0330657959,
63
- "avg_doclen_est": 171.25559997558594
64
  }
 
37
  "checkpoint": "colbert-ir\/colbertv2.0",
38
  "triples": "\/future\/u\/okhattab\/root\/unit\/experiments\/2021.10\/downstream.distillation.round2.2_score\/round2.nway6.cosine.ib\/examples.64.json",
39
  "collection": [
40
+ "list with 5498 elements starting with...",
41
  [
42
  "Driven by large-data pre-training, Segment Anything Model (SAM) has been demonstrated as a powerful and promptable framework, revolutionizing the segmentation models. Despite the generality, customizing SAM for specific visual concepts without man-powered prompting is under explored, e.g., automatically segmenting your pet dog in different images. In this paper, we propose a training-free Personalization approach for SAM, termed as PerSAM. Given only a single image with a reference mask, PerSAM first localizes the target concept by a location prior, and segments it within other images or videos via three techniques: target-guided attention, target-semantic prompting, and cascaded post-refinement. In this way, we effectively adapt SAM for private use without any training. To further alleviate the mask ambiguity, we present an efficient one-shot fine-tuning variant, PerSAM-F. Freezing the entire SAM, we introduce two learnable weights for multi-scale masks, only training 2 parameters within 10 seconds for improved performance. To demonstrate our efficacy, we construct a new segmentation dataset, PerSeg, for personalized evaluation, and test our methods on video object segmentation with competitive performance.",
43
  "Freezing the entire SAM, we introduce two learnable weights for multi-scale masks, only training 2 parameters within 10 seconds for improved performance. To demonstrate our efficacy, we construct a new segmentation dataset, PerSeg, for personalized evaluation, and test our methods on video object segmentation with competitive performance. Besides, our approach can also enhance DreamBooth to personalize Stable Diffusion for text-to-image generation, which discards the background disturbance for better target appearance learning. Code is released at https:\/\/github.com\/ZrrSkywalker\/Personalize-SAM",
 
50
  "root": ".ragatouille\/",
51
  "experiment": "colbert",
52
  "index_root": null,
53
+ "name": "2024-10\/09\/09.54.21",
54
  "rank": 0,
55
  "nranks": 1,
56
  "amp": true,
 
59
  },
60
  "num_chunks": 1,
61
  "num_partitions": 8192,
62
+ "num_embeddings_est": 941543.9933166504,
63
+ "avg_doclen_est": 171.25209045410156
64
  }