hysts-bot commited on
Commit
18b50ec
·
verified ·
1 Parent(s): f951133

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 sha256:5d0a3dcefb5ce7523d61a5e0fc8775ea7974d2d6cc4f9d04577ba21e5974796e
3
- size 3341788
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:64b7b51518f66f495e64f096d4f94fdcbd20b708c9b92aec3c5c74ee8a04f0c2
3
+ size 3342620
0.metadata.json CHANGED
@@ -1,6 +1,6 @@
1
  {
2
  "passage_offset": 0,
3
- "num_passages": 4877,
4
- "num_embeddings": 835164,
5
  "embedding_offset": 0
6
  }
 
1
  {
2
  "passage_offset": 0,
3
+ "num_passages": 4878,
4
+ "num_embeddings": 835360,
5
  "embedding_offset": 0
6
  }
0.residuals.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:1532c09f843e4546bcc2e7314e88be6b41932340b027627d89402b9a33eee93c
3
- size 106902192
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7bee5541841d27f28f01ff6efee12be22f79692f476a1fd409d27b85f7c867bb
3
+ size 106927280
avg_residual.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:85547f2aeba6b59b0b140a8dd9211dc9738c7c5ff590dab46093f362244c60c7
3
  size 1205
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f565abb23a3e7e08075ee2c3d3c744265e66bee7f2c4206b1ea7781b827882d7
3
  size 1205
buckets.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c6cfffb52c55884108ec6ee722c76ca6cba911d9e52d459ceb01f2e4fb4585ae
3
  size 2904
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:04dd74279e70208c8044cc488e9e479a0e83c8e30fa18725fcb4682098d73c7b
3
  size 2904
centroids.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:7511c5b4679d08d3979c33f01ea952fdac7d70222c3bcfe5225d9fa8f6f0801d
3
  size 2098342
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1390aeee2f82a429ab58eedd54742d48ffe89cb15e6048f8bfd2aee92875d16d
3
  size 2098342
collection.json CHANGED
@@ -4875,5 +4875,6 @@
4875
  "Secondly, we employ LMMs with the image generation tool to produce visual images of settings and scenes. These images are designed to maintain visual consistency across different scenes using a visual-language prompting method that combines scene descriptions and images of the appearing character and setting. Thirdly, scene images serve as the foundation for producing animated videos, with LMMs generating prompts to guide this process. The whole process is notably autonomous without manual intervention, as the LMMs interact seamlessly with generative tools to generate prompts, evaluate visual quality, and select the best one to optimize the final output.",
4876
  "Large language models (LLMs) have achieved impressive advancements across numerous disciplines, yet the critical issue of knowledge conflicts, a major source of hallucinations, has rarely been studied. Only a few research explored the conflicts between the inherent knowledge of LLMs and the retrieved contextual knowledge. However, a thorough assessment of knowledge conflict in LLMs is still missing. Motivated by this research gap, we present ConflictBank, the first comprehensive benchmark developed to systematically evaluate knowledge conflicts from three aspects: (i) conflicts encountered in retrieved knowledge, (ii) conflicts within the models' encoded knowledge, and (iii) the interplay between these conflict forms. Our investigation delves into four model families and twelve LLM instances, meticulously analyzing conflicts stemming from misinformation, temporal discrepancies, and semantic divergences. Based on our proposed novel construction framework, we create 7,453,853 claim-evidence pairs and 553,117 QA pairs. We present numerous findings on model scale, conflict causes, and conflict types. We hope our ConflictBank benchmark will help the community better understand model behavior in conflicts and develop more reliable LLMs.",
4877
  "In this report, we introduce Vintern-1B, a reliable 1-billion-parameters multimodal large language model (MLLM) for Vietnamese language tasks. By integrating the Qwen2-0.5B-Instruct language model with the InternViT-300M-448px visual model, Vintern-1B is optimized for a range of applications, including optical character recognition (OCR), document extraction, and general question-answering in Vietnamese context. The model is fine-tuned on an extensive dataset of over 3 million image-question-answer pairs, achieving robust performance and reliable results across multiple Vietnamese language benchmarks like OpenViVQA and ViTextVQA. Vintern-1B is small enough to fit into various on-device applications easily. Additionally, we have open-sourced several Vietnamese vision question answering (VQA) datasets for text and diagrams, created with Gemini 1.5 Flash. Our models are available at: https://huggingface.co/5CD-AI/Vintern-1B-v2.",
4878
- "Embedding models play a crucial role in Natural Language Processing (NLP) by creating text embeddings used in various tasks such as information retrieval and assessing semantic text similarity. This paper focuses on research related to embedding models in the Russian language. It introduces a new Russian-focused embedding model called ru-en-RoSBERTa and the ruMTEB benchmark, the Russian version extending the Massive Text Embedding Benchmark (MTEB). Our benchmark includes seven categories of tasks, such as semantic textual similarity, text classification, reranking, and retrieval. The research also assesses a representative set of Russian and multilingual models on the proposed benchmark. The findings indicate that the new model achieves results that are on par with state-of-the-art models in Russian. We release the model ru-en-RoSBERTa, and the ruMTEB framework comes with open-source code, integration into the original framework and a public leaderboard."
 
4879
  ]
 
4875
  "Secondly, we employ LMMs with the image generation tool to produce visual images of settings and scenes. These images are designed to maintain visual consistency across different scenes using a visual-language prompting method that combines scene descriptions and images of the appearing character and setting. Thirdly, scene images serve as the foundation for producing animated videos, with LMMs generating prompts to guide this process. The whole process is notably autonomous without manual intervention, as the LMMs interact seamlessly with generative tools to generate prompts, evaluate visual quality, and select the best one to optimize the final output.",
4876
  "Large language models (LLMs) have achieved impressive advancements across numerous disciplines, yet the critical issue of knowledge conflicts, a major source of hallucinations, has rarely been studied. Only a few research explored the conflicts between the inherent knowledge of LLMs and the retrieved contextual knowledge. However, a thorough assessment of knowledge conflict in LLMs is still missing. Motivated by this research gap, we present ConflictBank, the first comprehensive benchmark developed to systematically evaluate knowledge conflicts from three aspects: (i) conflicts encountered in retrieved knowledge, (ii) conflicts within the models' encoded knowledge, and (iii) the interplay between these conflict forms. Our investigation delves into four model families and twelve LLM instances, meticulously analyzing conflicts stemming from misinformation, temporal discrepancies, and semantic divergences. Based on our proposed novel construction framework, we create 7,453,853 claim-evidence pairs and 553,117 QA pairs. We present numerous findings on model scale, conflict causes, and conflict types. We hope our ConflictBank benchmark will help the community better understand model behavior in conflicts and develop more reliable LLMs.",
4877
  "In this report, we introduce Vintern-1B, a reliable 1-billion-parameters multimodal large language model (MLLM) for Vietnamese language tasks. By integrating the Qwen2-0.5B-Instruct language model with the InternViT-300M-448px visual model, Vintern-1B is optimized for a range of applications, including optical character recognition (OCR), document extraction, and general question-answering in Vietnamese context. The model is fine-tuned on an extensive dataset of over 3 million image-question-answer pairs, achieving robust performance and reliable results across multiple Vietnamese language benchmarks like OpenViVQA and ViTextVQA. Vintern-1B is small enough to fit into various on-device applications easily. Additionally, we have open-sourced several Vietnamese vision question answering (VQA) datasets for text and diagrams, created with Gemini 1.5 Flash. Our models are available at: https://huggingface.co/5CD-AI/Vintern-1B-v2.",
4878
+ "Embedding models play a crucial role in Natural Language Processing (NLP) by creating text embeddings used in various tasks such as information retrieval and assessing semantic text similarity. This paper focuses on research related to embedding models in the Russian language. It introduces a new Russian-focused embedding model called ru-en-RoSBERTa and the ruMTEB benchmark, the Russian version extending the Massive Text Embedding Benchmark (MTEB). Our benchmark includes seven categories of tasks, such as semantic textual similarity, text classification, reranking, and retrieval. The research also assesses a representative set of Russian and multilingual models on the proposed benchmark. The findings indicate that the new model achieves results that are on par with state-of-the-art models in Russian. We release the model ru-en-RoSBERTa, and the ruMTEB framework comes with open-source code, integration into the original framework and a public leaderboard.",
4879
+ "Given the widespread dissemination of misinformation on social media, implementing fact-checking mechanisms for online claims is essential. Manually verifying every claim is highly challenging, underscoring the need for an automated fact-checking system. This paper presents our system designed to address this issue. We utilize the Averitec dataset to assess the veracity of claims. In addition to veracity prediction, our system provides supporting evidence, which is extracted from the dataset. We develop a Retrieve and Generate (RAG) pipeline to extract relevant evidence sentences from a knowledge base, which are then inputted along with the claim into a large language model (LLM) for classification. We also evaluate the few-shot In-Context Learning (ICL) capabilities of multiple LLMs. Our system achieves an 'Averitec' score of 0.33, which is a 22% absolute improvement over the baseline. All code will be made available on All code will be made available on https://github.com/ronit-singhal/evidence-backed-fact-checking-using-rag-and-few-shot-in-context-learning-with-llms."
4880
  ]
doclens.0.json CHANGED
@@ -1 +1 @@
1
- [178,205,218,148,184,163,221,185,200,228,172,155,210,222,88,206,226,67,132,212,91,206,104,212,174,205,132,159,230,175,216,198,227,190,212,198,122,213,169,204,92,197,118,191,191,224,69,219,197,72,218,77,175,111,155,217,220,170,231,91,221,217,95,146,177,123,195,205,151,209,207,36,202,200,226,176,232,53,167,199,89,184,213,104,154,153,216,214,215,174,205,72,211,78,221,212,232,223,73,158,220,158,202,222,189,165,205,175,222,132,126,179,219,110,209,158,208,98,176,192,226,34,158,205,126,178,224,182,227,100,152,191,169,195,163,172,208,117,199,217,167,217,157,163,194,217,200,217,23,221,209,146,150,204,200,125,215,232,68,147,212,41,223,178,152,173,210,139,198,182,196,207,95,176,205,223,83,216,85,207,210,52,177,178,230,197,119,226,99,182,210,212,77,138,199,123,179,111,219,69,223,65,204,215,83,197,87,211,132,216,135,178,157,166,216,85,170,195,208,190,175,134,220,200,67,221,211,66,227,222,226,190,209,205,67,207,139,208,127,186,205,168,221,179,223,117,148,221,216,80,189,125,199,202,64,218,77,195,190,221,181,98,143,214,220,97,187,127,219,122,216,87,138,212,194,112,219,227,101,220,100,164,234,109,221,102,223,89,184,205,219,77,188,223,172,171,175,152,175,137,213,197,114,205,221,138,181,174,227,73,147,144,178,147,215,152,182,204,80,210,123,211,121,209,224,224,219,211,163,133,187,148,151,163,221,94,133,213,72,187,224,216,162,154,224,184,118,204,220,154,117,220,162,202,223,195,110,197,151,224,88,182,217,221,214,118,218,118,164,205,97,221,183,154,206,197,74,170,219,103,230,215,192,224,78,184,72,201,227,221,191,181,104,190,224,221,99,123,206,102,228,202,74,195,96,225,176,232,231,96,225,206,61,173,101,190,211,90,213,199,203,184,209,173,160,207,203,99,201,195,132,195,214,148,211,227,192,215,212,127,162,213,114,178,111,207,71,129,182,212,75,176,209,137,213,224,220,70,232,116,228,179,200,184,222,93,202,71,219,123,213,119,204,173,135,118,207,96,216,107,210,202,189,172,211,76,146,195,169,83,227,62,219,223,125,158,202,226,98,214,225,76,156,211,204,110,190,224,200,64,228,223,72,213,102,214,178,219,128,211,127,187,213,160,216,134,202,150,186,201,195,230,199,206,193,205,219,216,84,213,223,192,222,109,187,209,79,153,204,224,72,167,190,183,183,228,133,227,103,197,225,203,205,201,207,213,210,78,138,195,133,195,180,190,185,207,189,100,191,229,198,129,226,135,205,164,226,190,73,200,231,209,216,147,221,228,165,213,209,181,157,222,117,216,58,148,219,92,222,219,63,191,218,186,197,158,172,204,109,204,102,223,229,91,192,217,186,204,144,173,209,95,186,222,225,178,183,222,71,193,182,232,63,227,228,82,209,162,219,183,222,122,212,195,60,219,130,223,69,216,111,198,224,80,223,74,164,148,227,193,216,139,207,198,148,227,199,169,230,225,86,138,165,206,91,213,164,215,229,173,216,210,92,166,205,102,196,78,202,74,187,202,229,191,222,57,206,115,183,94,225,173,198,109,212,224,126,215,44,221,80,196,95,177,218,217,198,109,218,191,221,214,227,71,207,123,206,90,219,218,108,181,214,123,150,204,214,208,225,199,191,192,213,93,217,149,178,199,106,204,80,208,60,131,132,232,218,213,115,202,94,175,227,217,114,200,222,178,180,185,192,193,205,173,226,217,222,197,184,205,211,136,214,221,62,213,89,169,184,135,198,98,216,225,71,230,133,213,154,216,189,122,224,108,204,88,207,227,93,191,204,214,66,153,219,209,203,218,237,229,222,148,174,218,107,214,169,223,216,173,220,148,199,188,175,203,142,188,210,98,167,174,224,103,156,97,220,111,216,70,130,221,216,115,166,164,162,210,216,69,173,203,77,216,175,172,210,98,199,213,88,221,82,184,131,201,220,133,211,116,156,219,213,227,128,221,237,172,217,64,180,221,70,169,220,216,134,215,177,190,197,229,221,226,74,189,190,219,86,222,74,213,106,184,182,205,221,229,143,201,198,162,209,128,221,209,60,216,146,228,106,215,154,214,202,82,194,141,168,217,213,80,140,114,220,149,229,215,150,211,170,199,109,218,108,206,125,204,164,183,148,193,153,188,143,160,199,210,124,225,105,178,75,222,112,205,75,199,83,222,215,158,104,205,70,215,223,158,220,99,124,194,91,219,65,209,149,234,188,222,218,200,66,208,101,208,188,201,207,198,160,205,219,118,170,195,141,128,182,189,211,51,179,216,112,152,196,224,80,223,177,218,127,186,113,151,117,159,184,200,231,212,134,222,68,216,204,90,151,202,158,173,146,144,209,222,89,171,188,206,215,117,191,223,80,215,68,151,159,210,74,208,105,204,105,174,228,67,147,202,214,41,222,163,183,142,227,188,200,208,132,157,191,195,94,166,201,155,214,132,185,195,165,214,219,156,217,100,147,220,105,220,223,225,117,176,166,207,134,218,202,172,183,217,76,229,139,217,87,208,181,214,169,154,199,231,208,200,157,213,222,49,207,191,139,219,218,82,122,154,137,190,176,192,180,231,80,208,136,185,112,189,187,192,208,219,110,208,102,186,225,220,218,235,214,96,133,198,199,95,152,83,219,190,124,148,222,213,207,145,217,168,200,32,208,198,171,205,218,104,218,136,230,77,139,178,212,214,129,211,233,93,221,95,155,153,221,151,201,108,193,216,172,208,72,177,217,208,225,60,81,172,234,180,146,184,180,199,222,224,182,106,183,105,221,208,81,221,137,211,44,198,224,205,82,187,164,147,213,163,229,192,219,218,190,228,221,218,116,231,97,182,126,213,193,173,207,103,207,122,161,156,193,153,192,92,198,155,223,224,106,223,170,177,225,40,206,203,221,152,208,116,203,105,197,221,65,226,111,224,215,215,142,154,211,84,184,135,169,212,213,76,229,225,130,207,85,209,177,116,140,150,229,209,105,164,186,214,219,62,148,203,230,118,207,216,217,211,114,201,134,217,87,197,124,209,186,184,164,189,163,212,85,219,106,209,220,168,221,212,112,189,183,229,200,192,106,224,114,220,95,141,223,222,219,226,79,187,202,212,222,193,128,223,209,160,178,197,222,111,152,114,164,206,179,189,183,122,212,149,150,130,214,115,195,118,204,208,120,195,163,159,220,125,222,174,211,144,140,220,80,173,204,87,215,149,227,97,229,221,119,213,139,212,82,215,68,213,136,186,168,202,85,213,93,168,218,63,222,121,213,90,218,94,197,221,182,220,112,224,221,83,215,194,138,181,197,221,91,186,194,99,179,213,210,76,188,214,216,79,213,92,220,94,201,108,213,131,221,153,219,74,159,172,218,217,219,108,209,35,217,112,157,206,148,211,126,214,107,191,105,164,224,220,126,204,201,211,119,203,129,165,205,130,218,137,211,178,204,215,206,119,209,75,193,182,115,178,207,217,220,184,199,167,189,230,140,208,167,173,195,136,220,205,64,186,159,194,75,213,189,110,189,208,120,176,182,162,196,183,137,134,198,172,193,184,172,212,189,214,200,95,167,199,110,195,116,180,204,66,217,126,222,226,193,166,143,121,218,224,196,140,226,142,206,66,214,140,218,144,201,221,152,204,154,221,101,225,69,216,212,128,181,200,180,206,136,193,226,226,219,141,168,202,209,105,178,128,208,197,145,163,218,124,220,147,220,156,211,69,212,226,132,223,165,211,110,229,194,193,204,208,90,218,106,152,215,134,180,182,204,211,80,211,108,204,102,208,102,192,94,191,206,203,140,165,213,197,105,186,186,210,62,208,165,213,174,186,115,211,164,193,128,201,169,217,167,217,174,222,167,129,170,217,218,182,159,113,208,85,219,126,219,89,172,213,212,69,205,135,124,205,67,208,52,186,160,225,125,190,65,227,168,180,166,151,194,205,88,138,204,110,124,223,80,193,106,179,183,163,220,148,205,218,200,221,198,202,190,213,87,227,239,81,213,115,211,219,201,204,105,216,83,210,95,170,219,197,86,228,86,209,125,231,83,198,194,171,144,183,222,216,216,226,204,109,216,115,180,227,116,207,223,84,190,215,117,212,79,214,230,218,134,208,92,209,99,181,220,61,212,85,199,185,174,171,220,221,57,146,224,214,81,183,218,152,163,193,72,203,111,130,222,204,217,229,209,201,189,220,213,72,192,214,221,157,197,124,209,100,223,162,223,73,85,221,68,210,46,205,76,118,160,180,200,85,218,34,196,170,204,88,168,224,71,188,207,166,181,210,101,148,215,76,219,217,104,199,180,210,208,69,201,202,216,224,89,206,88,128,125,196,203,106,183,182,84,228,200,120,171,166,153,215,210,82,210,65,163,131,213,83,180,205,92,152,228,99,212,212,209,94,215,218,95,206,179,236,219,204,102,196,110,160,218,140,228,102,218,177,172,193,95,209,215,203,220,78,213,191,189,168,204,218,192,89,213,223,208,230,168,202,169,222,56,155,203,73,197,156,208,222,127,212,223,160,217,201,223,205,217,100,221,208,143,218,22,208,224,86,209,90,220,144,182,139,219,79,194,183,225,216,209,190,102,221,220,203,152,215,113,230,106,212,221,106,197,111,185,211,83,162,200,208,219,124,210,92,161,216,221,79,208,166,211,142,179,222,197,79,185,226,182,193,195,170,202,113,192,231,228,197,180,152,223,155,215,97,213,85,205,145,200,89,209,201,139,217,163,133,206,174,112,205,213,217,123,201,136,132,221,206,105,200,139,183,160,215,134,221,208,181,208,101,210,225,116,183,217,80,186,120,162,194,201,97,188,209,174,122,229,214,205,120,218,228,228,83,134,201,119,171,210,55,227,107,225,64,174,203,130,213,226,95,206,157,139,154,171,139,224,219,154,190,206,218,217,221,86,135,217,213,183,118,187,186,128,212,79,184,186,156,188,134,216,172,155,195,210,176,223,125,128,217,213,214,217,94,189,192,133,167,85,187,217,60,158,221,122,217,117,224,95,206,89,199,74,205,210,70,209,83,196,77,169,188,228,114,188,105,229,182,174,223,59,181,205,127,225,222,128,190,219,112,197,115,222,205,115,184,214,210,238,196,212,85,207,86,183,215,178,187,112,194,176,212,90,217,80,164,179,187,138,221,214,203,96,227,198,82,209,207,96,208,76,228,224,205,207,186,178,221,224,189,143,229,182,229,132,213,201,186,218,67,222,221,83,216,44,220,68,214,103,137,210,48,210,211,88,222,169,225,155,181,200,207,62,219,152,179,130,215,86,183,90,202,164,179,205,128,219,224,148,190,117,213,221,184,175,146,223,216,82,230,92,201,231,218,172,207,209,189,150,216,142,226,46,197,219,92,228,168,216,111,221,204,216,190,180,189,206,108,204,101,224,74,223,65,181,142,211,203,122,220,214,66,179,185,222,116,120,209,222,101,172,209,232,56,223,177,212,107,191,168,193,193,105,212,105,163,209,191,117,194,170,211,190,121,221,90,161,207,67,226,90,216,60,171,219,89,220,184,214,194,144,182,214,208,196,133,226,214,77,190,215,63,115,208,92,227,204,68,197,77,178,188,204,216,202,194,152,175,181,208,131,196,223,128,177,116,207,102,220,121,216,199,202,207,74,211,196,217,177,214,218,221,93,219,192,215,143,139,183,226,83,102,219,211,100,194,47,119,204,160,143,180,149,183,226,96,172,212,186,198,207,182,207,221,164,146,224,61,200,140,147,196,214,60,226,215,111,216,186,180,161,234,196,124,208,214,221,189,178,112,168,208,136,169,181,202,124,201,74,211,143,192,175,119,163,202,67,191,172,148,179,223,122,217,117,180,190,186,209,72,193,182,218,80,156,138,201,202,218,208,149,190,109,224,64,192,114,219,199,210,208,97,232,205,197,188,189,87,213,152,196,164,131,225,94,219,205,163,191,172,196,189,206,110,201,73,191,122,210,173,208,73,221,136,222,185,224,74,213,86,185,221,170,210,69,165,156,210,102,211,210,221,197,209,154,127,212,180,208,139,231,207,50,166,96,150,158,206,228,214,202,95,162,220,191,136,217,54,155,201,140,179,191,91,213,220,74,145,216,232,45,208,217,209,182,160,182,100,221,155,219,227,160,180,209,147,174,212,67,209,82,213,148,208,51,176,88,210,71,117,174,218,82,211,188,170,210,186,136,176,220,157,189,167,190,212,160,212,135,201,219,49,162,165,209,117,175,213,152,176,220,221,124,150,204,220,73,228,194,218,57,195,173,159,173,175,206,176,229,79,164,201,203,152,214,116,137,219,66,222,214,80,175,202,90,225,113,219,206,78,190,89,214,50,209,154,188,227,194,157,195,74,186,206,130,198,73,212,60,204,122,222,99,205,196,229,213,83,230,108,171,126,192,104,216,207,117,217,197,214,79,207,110,221,79,217,144,206,160,206,172,197,183,207,217,207,113,210,221,71,161,221,164,227,214,142,177,185,180,103,130,198,123,205,74,216,102,219,160,217,75,204,114,192,213,166,188,118,222,227,92,195,219,161,200,221,69,203,143,198,198,217,198,66,212,50,208,116,199,125,210,207,167,225,116,207,97,184,99,220,184,203,184,219,177,167,202,214,55,207,161,197,122,212,226,187,96,216,201,188,135,224,207,139,225,230,220,121,221,107,212,66,170,169,210,199,102,220,94,159,184,207,92,207,231,214,125,227,220,205,58,193,203,215,223,229,78,196,170,185,196,162,234,56,201,123,171,231,196,86,162,199,213,220,68,200,68,205,88,225,135,220,82,182,215,222,79,152,230,62,162,218,184,224,67,206,99,189,124,214,197,73,204,105,221,179,102,218,232,80,214,181,170,204,165,216,207,217,212,195,176,215,106,192,160,221,182,217,57,211,88,198,233,113,171,204,138,193,209,225,59,176,184,134,223,151,193,200,217,100,225,79,180,142,190,123,222,80,232,216,133,216,148,211,110,198,96,187,224,95,208,112,178,227,94,171,96,181,209,170,225,196,206,94,216,87,217,171,191,82,218,127,227,176,219,207,230,79,214,203,105,213,143,174,188,125,193,220,60,215,172,214,101,211,110,161,117,187,180,125,218,220,62,208,203,217,87,198,156,216,226,161,161,223,224,72,178,198,213,195,219,208,140,175,217,74,201,201,66,186,154,229,89,226,169,204,87,184,85,161,133,201,80,176,188,114,224,77,207,126,202,83,219,200,125,172,169,190,216,80,88,221,68,218,133,216,117,217,157,217,170,190,124,214,210,156,231,84,207,204,113,200,70,222,162,208,227,92,223,136,167,195,221,221,77,173,213,109,214,117,211,217,89,217,91,210,152,194,206,202,110,216,177,190,207,227,185,172,230,172,207,171,199,234,207,149,194,192,179,212,209,210,101,198,225,85,164,211,110,194,182,211,224,65,228,218,79,224,81,122,208,154,129,206,92,193,171,148,188,221,80,220,161,165,166,161,214,99,210,64,174,224,221,105,200,122,230,216,94,223,128,225,161,219,126,187,137,191,222,214,148,151,198,218,210,110,208,228,184,211,35,202,218,195,216,115,212,95,177,199,101,184,208,202,212,134,193,129,192,81,182,223,70,226,230,134,167,183,198,222,227,227,226,63,213,109,187,177,219,223,203,144,179,209,103,177,181,158,221,90,222,166,207,175,230,207,99,205,234,210,210,168,223,143,210,187,209,204,150,209,213,208,193,221,214,77,215,199,81,197,82,177,190,210,231,79,179,221,64,182,199,82,204,204,95,172,187,178,209,86,222,220,118,192,223,88,220,77,174,104,224,137,182,186,96,207,198,74,152,196,217,206,79,214,208,204,180,94,215,81,177,160,201,164,173,205,76,199,220,228,91,215,155,226,79,133,181,136,182,226,96,221,109,209,223,71,202,95,217,87,202,204,183,210,187,212,81,226,184,224,88,170,214,198,226,142,212,81,209,189,172,192,221,216,123,221,126,204,218,222,76,205,73,225,221,73,204,108,201,88,174,197,136,223,90,189,56,207,147,206,212,73,201,83,204,112,137,227,67,208,137,219,225,65,200,186,99,214,97,215,74,203,65,199,216,108,216,80,206,219,104,226,180,225,199,186,197,226,157,102,177,107,231,156,141,226,70,220,216,223,64,214,66,201,174,170,207,46,202,131,173,218,125,217,157,234,192,159,174,209,95,196,224,59,220,69,211,130,203,222,88,208,86,198,127,219,228,75,218,170,168,198,128,215,54,211,167,186,117,211,162,221,219,105,223,99,223,127,202,218,213,143,194,181,200,180,230,224,97,181,132,173,202,221,57,151,220,77,220,160,206,188,101,197,72,213,95,193,212,189,105,226,100,205,201,56,211,93,178,212,88,208,83,213,165,219,183,236,121,220,210,94,212,171,186,218,137,212,129,175,203,223,134,194,95,193,191,105,229,208,102,196,120,191,221,217,65,206,200,74,168,180,199,217,119,223,68,211,125,204,105,180,164,215,227,128,211,166,218,86,185,74,214,57,200,171,111,185,73,199,220,213,192,216,107,211,115,219,227,192,221,101,203,65,211,51,216,84,193,121,214,86,195,115,179,229,90,215,92,207,63,179,212,38,202,104,182,125,179,99,147,184,210,166,227,232,164,120,218,169,203,154,192,224,217,122,160,205,206,221,80,191,217,166,202,78,206,147,202,155,195,76,204,136,191,112,195,160,147,226,91,224,216,212,177,188,165,174,130,203,221,220,133,209,147,216,69,159,155,143,213,94,227,139,209,163,183,199,112,217,213,98,217,96,185,158,173,229,51,209,195,227,214,161,213,83,168,229,209,118,221,224,59,179,161,220,209,193,199,199,212,107,226,219,204,117,166,223,122,166,181,163,176,223,176,130,223,221,202,89,188,147,160,143,218,223,206,151,201,161,130,176,175,138,126,209,112,230,94,211,17,103,218,73,218,131,210,104,214,63,222,38,135,140,215,143,215,191,185,223,207,215,203,46,219,207,93,177,85,213,191,223,56,181,209,82,210,221,66,210,195,223,184,138,217,48,194,73,150,199,220,183,209,60,194,103,218,103,211,216,124,197,217,185,106,185,207,174,165,204,138,220,68,218,151,202,68,214,155,183,221,66,216,61,218,122,214,178,202,178,217,142,215,126,187,148,219,98,180,222,217,80,210,203,43,208,154,220,101,167,206,211,212,208,72,147,225,139,174,207,36,200,234,205,211,180,205,202,126,159,186,116,211,154,192,155,194,168,198,160,218,220,202,153,222,215,66,174,128,211,104,136,171,235,219,112,156,209,109,203,132,192,181,215,112,205,68,215,82,213,117,189,221,186,211,171,208,136,189,128,210,96,199,107,195,232,74,223,132,193,198,46,220,73,181,112,224,133,221,144,224,83,232,217,131,186,53,214,225,95,203,70,102,217,106,224,79,210,113,177,150,228,220,102,225,80,221,170,206,105,223,112,210,46,201,89,197,207,128,235,111,212,161,144,221,182,200,77,213,229,90,134,223,179,212,204,125,197,215,80,233,218,44,226,53,152,184,220,113,219,216,110,214,206,151,215,224,216,163,144,190,133,223,195,216,203,67,95,169,191,131,208,78,104,176,179,148,207,172,220,98,202,118,218,204,120,213,92,213,93,210,203,219,75,212,227,212,188,187,201,100,206,151,200,96,197,215,157,210,70,207,182,205,205,101,212,117,230,86,163,143,167,189,215,168,216,194,98,218,128,219,94,149,188,217,48,172,174,131,131,182,171,200,115,220,217,91,200,80,178,188,226,49,192,205,222,127,194,134,175,214,115,212,214,82,193,106,217,48,197,114,204,114,201,221,190,174,214,168,223,86,217,212,214,181,204,96,190,189,216,91,204,118,226,110,198,224,158,195,117,189,200,150,231,206,91,209,207,118,223,183,232,146,158,207,213,122,204,118,200,103,200,227,76,179,195,73,215,93,214,170,215,232,41,210,107,138,202,204,125,198,134,225,80,117,164,185,197,106,232,74,139,216,207,209,57,207,94,200,229,190,192,140,112,208,155,191,177,216,203,142,192,103,195,219,91,179,228,187,115,213,217,192,193,215,141,218,70,186,37,225,190,84,178,177,162,218,210,185,176,195,97,218,63,218,176,227,215,149,224,221,115,208,214,133,182,188,205,163,207,58,199,217,129,208,184,214,206,133,228,200,80,224,179,229,152,208,95,194,170,224,178,196,181,94,199,90,203,75,216,153,169,213,86,225,67,192,194,65,189,201,103,188,153,163,213,226,142,185,133,226,106,181,215,199,209,211,56,204,114,181,163,171,228,228,73,74,198,203,186,178,185,125,229,221,204,41,165,189,126,205,173,116,179,198,159,216,129,209,222,174,183,229,95,212,68,177,152,217,138,156,135,105,204,193,188,127,209,150,133,163,208,80,218,232,211,77,230,37,198,224,165,213,81,220,207,195,173,174,212,102,206,117,196,178,222,134,205,216,203,86,151,184,157,217,222,123,213,159,195,121,207,159,220,120,212,176,144,217,67,173,216,105,206,90,220,121,217,59,184,219,156,213,149,143,216,221,140,225,182,115,209,115,209,45,223,172,227,133,187,165,199,168,224,91,206,115,153,202,197,62,169,210,134,215,167,203,214,141,200,213,182,90,214,170,206,199,219,54,167,154,72,194,122,181,197,129,214,105,153,209,137,202,227,72,207,61,178,127,181,210,200,46,188,210,214,67,189,216,51,209,125,190,127,208,110,191,219,137,213,76,206,120,186,121,201,222,113,195,194,68,183,179,184,223,61,180,220,197,133,208,226,136,217,200,93,178,220,113,197,198,172,141,225,102,159,149,213,196,100,220,196,176,232,182,187,171,165,182,101,175,169,191,224,110,200,128,200,129,114,179,188,165,198,216,184,174,216,67,229,198,220,32,232,219,72,219,203,127,88,212,81,142,223,210,166,97,145,209,77,216,227,196,83,202,137,214,82,223,114,205,177,183]
 
1
+ [178,205,218,148,184,163,221,185,200,228,172,155,210,222,88,206,226,67,132,212,91,206,104,212,174,205,132,159,230,175,216,198,227,190,212,198,122,213,169,204,92,197,118,191,191,224,69,219,197,72,218,77,175,111,155,217,220,170,231,91,221,217,95,146,177,123,195,205,151,209,207,36,202,200,226,176,232,53,167,199,89,184,213,104,154,153,216,214,215,174,205,72,211,78,221,212,232,223,73,158,220,158,202,222,189,165,205,175,222,132,126,179,219,110,209,158,208,98,176,192,226,34,158,205,126,178,224,182,227,100,152,191,169,195,163,172,208,117,199,217,167,217,157,163,194,217,200,217,23,221,209,146,150,204,200,125,215,232,68,147,212,41,223,178,152,173,210,139,198,182,196,207,95,176,205,223,83,216,85,207,210,52,177,178,230,197,119,226,99,182,210,212,77,138,199,123,179,111,219,69,223,65,204,215,83,197,87,211,132,216,135,178,157,166,216,85,170,195,208,190,175,134,220,200,67,221,211,66,227,222,226,190,209,205,67,207,139,208,127,186,205,168,221,179,223,117,148,221,216,80,189,125,199,202,64,218,77,195,190,221,181,98,143,214,220,97,187,127,219,122,216,87,138,212,194,112,219,227,101,220,100,164,234,109,221,102,223,89,184,205,219,77,188,223,172,171,175,152,175,137,213,197,114,205,221,138,181,174,227,73,147,144,178,147,215,152,182,204,80,210,123,211,121,209,224,224,219,211,163,133,187,148,151,163,221,94,133,213,72,187,224,216,162,154,224,184,118,204,220,154,117,220,162,202,223,195,110,197,151,224,88,182,217,221,214,118,218,118,164,205,97,221,183,154,206,197,74,170,219,103,230,215,192,224,78,184,72,201,227,221,191,181,104,190,224,221,99,123,206,102,228,202,74,195,96,225,176,232,231,96,225,206,61,173,101,190,211,90,213,199,203,184,209,173,160,207,203,99,201,195,132,195,214,148,211,227,192,215,212,127,162,213,114,178,111,207,71,129,182,212,75,176,209,137,213,224,220,70,232,116,228,179,200,184,222,93,202,71,219,123,213,119,204,173,135,118,207,96,216,107,210,202,189,172,211,76,146,195,169,83,227,62,219,223,125,158,202,226,98,214,225,76,156,211,204,110,190,224,200,64,228,223,72,213,102,214,178,219,128,211,127,187,213,160,216,134,202,150,186,201,195,230,199,206,193,205,219,216,84,213,223,192,222,109,187,209,79,153,204,224,72,167,190,183,183,228,133,227,103,197,225,203,205,201,207,213,210,78,138,195,133,195,180,190,185,207,189,100,191,229,198,129,226,135,205,164,226,190,73,200,231,209,216,147,221,228,165,213,209,181,157,222,117,216,58,148,219,92,222,219,63,191,218,186,197,158,172,204,109,204,102,223,229,91,192,217,186,204,144,173,209,95,186,222,225,178,183,222,71,193,182,232,63,227,228,82,209,162,219,183,222,122,212,195,60,219,130,223,69,216,111,198,224,80,223,74,164,148,227,193,216,139,207,198,148,227,199,169,230,225,86,138,165,206,91,213,164,215,229,173,216,210,92,166,205,102,196,78,202,74,187,202,229,191,222,57,206,115,183,94,225,173,198,109,212,224,126,215,44,221,80,196,95,177,218,217,198,109,218,191,221,214,227,71,207,123,206,90,219,218,108,181,214,123,150,204,214,208,225,199,191,192,213,93,217,149,178,199,106,204,80,208,60,131,132,232,218,213,115,202,94,175,227,217,114,200,222,178,180,185,192,193,205,173,226,217,222,197,184,205,211,136,214,221,62,213,89,169,184,135,198,98,216,225,71,230,133,213,154,216,189,122,224,108,204,88,207,227,93,191,204,214,66,153,219,209,203,218,237,229,222,148,174,218,107,214,169,223,216,173,220,148,199,188,175,203,142,188,210,98,167,174,224,103,156,97,220,111,216,70,130,221,216,115,166,164,162,210,216,69,173,203,77,216,175,172,210,98,199,213,88,221,82,184,131,201,220,133,211,116,156,219,213,227,128,221,237,172,217,64,180,221,70,169,220,216,134,215,177,190,197,229,221,226,74,189,190,219,86,222,74,213,106,184,182,205,221,229,143,201,198,162,209,128,221,209,60,216,146,228,106,215,154,214,202,82,194,141,168,217,213,80,140,114,220,149,229,215,150,211,170,199,109,218,108,206,125,204,164,183,148,193,153,188,143,160,199,210,124,225,105,178,75,222,112,205,75,199,83,222,215,158,104,205,70,215,223,158,220,99,124,194,91,219,65,209,149,234,188,222,218,200,66,208,101,208,188,201,207,198,160,205,219,118,170,195,141,128,182,189,211,51,179,216,112,152,196,224,80,223,177,218,127,186,113,151,117,159,184,200,231,212,134,222,68,216,204,90,151,202,158,173,146,144,209,222,89,171,188,206,215,117,191,223,80,215,68,151,159,210,74,208,105,204,105,174,228,67,147,202,214,41,222,163,183,142,227,188,200,208,132,157,191,195,94,166,201,155,214,132,185,195,165,214,219,156,217,100,147,220,105,220,223,225,117,176,166,207,134,218,202,172,183,217,76,229,139,217,87,208,181,214,169,154,199,231,208,200,157,213,222,49,207,191,139,219,218,82,122,154,137,190,176,192,180,231,80,208,136,185,112,189,187,192,208,219,110,208,102,186,225,220,218,235,214,96,133,198,199,95,152,83,219,190,124,148,222,213,207,145,217,168,200,32,208,198,171,205,218,104,218,136,230,77,139,178,212,214,129,211,233,93,221,95,155,153,221,151,201,108,193,216,172,208,72,177,217,208,225,60,81,172,234,180,146,184,180,199,222,224,182,106,183,105,221,208,81,221,137,211,44,198,224,205,82,187,164,147,213,163,229,192,219,218,190,228,221,218,116,231,97,182,126,213,193,173,207,103,207,122,161,156,193,153,192,92,198,155,223,224,106,223,170,177,225,40,206,203,221,152,208,116,203,105,197,221,65,226,111,224,215,215,142,154,211,84,184,135,169,212,213,76,229,225,130,207,85,209,177,116,140,150,229,209,105,164,186,214,219,62,148,203,230,118,207,216,217,211,114,201,134,217,87,197,124,209,186,184,164,189,163,212,85,219,106,209,220,168,221,212,112,189,183,229,200,192,106,224,114,220,95,141,223,222,219,226,79,187,202,212,222,193,128,223,209,160,178,197,222,111,152,114,164,206,179,189,183,122,212,149,150,130,214,115,195,118,204,208,120,195,163,159,220,125,222,174,211,144,140,220,80,173,204,87,215,149,227,97,229,221,119,213,139,212,82,215,68,213,136,186,168,202,85,213,93,168,218,63,222,121,213,90,218,94,197,221,182,220,112,224,221,83,215,194,138,181,197,221,91,186,194,99,179,213,210,76,188,214,216,79,213,92,220,94,201,108,213,131,221,153,219,74,159,172,218,217,219,108,209,35,217,112,157,206,148,211,126,214,107,191,105,164,224,220,126,204,201,211,119,203,129,165,205,130,218,137,211,178,204,215,206,119,209,75,193,182,115,178,207,217,220,184,199,167,189,230,140,208,167,173,195,136,220,205,64,186,159,194,75,213,189,110,189,208,120,176,182,162,196,183,137,134,198,172,193,184,172,212,189,214,200,95,167,199,110,195,116,180,204,66,217,126,222,226,193,166,143,121,218,224,196,140,226,142,206,66,214,140,218,144,201,221,152,204,154,221,101,225,69,216,212,128,181,200,180,206,136,193,226,226,219,141,168,202,209,105,178,128,208,197,145,163,218,124,220,147,220,156,211,69,212,226,132,223,165,211,110,229,194,193,204,208,90,218,106,152,215,134,180,182,204,211,80,211,108,204,102,208,102,192,94,191,206,203,140,165,213,197,105,186,186,210,62,208,165,213,174,186,115,211,164,193,128,201,169,217,167,217,174,222,167,129,170,217,218,182,159,113,208,85,219,126,219,89,172,213,212,69,205,135,124,205,67,208,52,186,160,225,125,190,65,227,168,180,166,151,194,205,88,138,204,110,124,223,80,193,106,179,183,163,220,148,205,218,200,221,198,202,190,213,87,227,239,81,213,115,211,219,201,204,105,216,83,210,95,170,219,197,86,228,86,209,125,231,83,198,194,171,144,183,222,216,216,226,204,109,216,115,180,227,116,207,223,84,190,215,117,212,79,214,230,218,134,208,92,209,99,181,220,61,212,85,199,185,174,171,220,221,57,146,224,214,81,183,218,152,163,193,72,203,111,130,222,204,217,229,209,201,189,220,213,72,192,214,221,157,197,124,209,100,223,162,223,73,85,221,68,210,46,205,76,118,160,180,200,85,218,34,196,170,204,88,168,224,71,188,207,166,181,210,101,148,215,76,219,217,104,199,180,210,208,69,201,202,216,224,89,206,88,128,125,196,203,106,183,182,84,228,200,120,171,166,153,215,210,82,210,65,163,131,213,83,180,205,92,152,228,99,212,212,209,94,215,218,95,206,179,236,219,204,102,196,110,160,218,140,228,102,218,177,172,193,95,209,215,203,220,78,213,191,189,168,204,218,192,89,213,223,208,230,168,202,169,222,56,155,203,73,197,156,208,222,127,212,223,160,217,201,223,205,217,100,221,208,143,218,22,208,224,86,209,90,220,144,182,139,219,79,194,183,225,216,209,190,102,221,220,203,152,215,113,230,106,212,221,106,197,111,185,211,83,162,200,208,219,124,210,92,161,216,221,79,208,166,211,142,179,222,197,79,185,226,182,193,195,170,202,113,192,231,228,197,180,152,223,155,215,97,213,85,205,145,200,89,209,201,139,217,163,133,206,174,112,205,213,217,123,201,136,132,221,206,105,200,139,183,160,215,134,221,208,181,208,101,210,225,116,183,217,80,186,120,162,194,201,97,188,209,174,122,229,214,205,120,218,228,228,83,134,201,119,171,210,55,227,107,225,64,174,203,130,213,226,95,206,157,139,154,171,139,224,219,154,190,206,218,217,221,86,135,217,213,183,118,187,186,128,212,79,184,186,156,188,134,216,172,155,195,210,176,223,125,128,217,213,214,217,94,189,192,133,167,85,187,217,60,158,221,122,217,117,224,95,206,89,199,74,205,210,70,209,83,196,77,169,188,228,114,188,105,229,182,174,223,59,181,205,127,225,222,128,190,219,112,197,115,222,205,115,184,214,210,238,196,212,85,207,86,183,215,178,187,112,194,176,212,90,217,80,164,179,187,138,221,214,203,96,227,198,82,209,207,96,208,76,228,224,205,207,186,178,221,224,189,143,229,182,229,132,213,201,186,218,67,222,221,83,216,44,220,68,214,103,137,210,48,210,211,88,222,169,225,155,181,200,207,62,219,152,179,130,215,86,183,90,202,164,179,205,128,219,224,148,190,117,213,221,184,175,146,223,216,82,230,92,201,231,218,172,207,209,189,150,216,142,226,46,197,219,92,228,168,216,111,221,204,216,190,180,189,206,108,204,101,224,74,223,65,181,142,211,203,122,220,214,66,179,185,222,116,120,209,222,101,172,209,232,56,223,177,212,107,191,168,193,193,105,212,105,163,209,191,117,194,170,211,190,121,221,90,161,207,67,226,90,216,60,171,219,89,220,184,214,194,144,182,214,208,196,133,226,214,77,190,215,63,115,208,92,227,204,68,197,77,178,188,204,216,202,194,152,175,181,208,131,196,223,128,177,116,207,102,220,121,216,199,202,207,74,211,196,217,177,214,218,221,93,219,192,215,143,139,183,226,83,102,219,211,100,194,47,119,204,160,143,180,149,183,226,96,172,212,186,198,207,182,207,221,164,146,224,61,200,140,147,196,214,60,226,215,111,216,186,180,161,234,196,124,208,214,221,189,178,112,168,208,136,169,181,202,124,201,74,211,143,192,175,119,163,202,67,191,172,148,179,223,122,217,117,180,190,186,209,72,193,182,218,80,156,138,201,202,218,208,149,190,109,224,64,192,114,219,199,210,208,97,232,205,197,188,189,87,213,152,196,164,131,225,94,219,205,163,191,172,196,189,206,110,201,73,191,122,210,173,208,73,221,136,222,185,224,74,213,86,185,221,170,210,69,165,156,210,102,211,210,221,197,209,154,127,212,180,208,139,231,207,50,166,96,150,158,206,228,214,202,95,162,220,191,136,217,54,155,201,140,179,191,91,213,220,74,145,216,232,45,208,217,209,182,160,182,100,221,155,219,227,160,180,209,147,174,212,67,209,82,213,148,208,51,176,88,210,71,117,174,218,82,211,188,170,210,186,136,176,220,157,189,167,190,212,160,212,135,201,219,49,162,165,209,117,175,213,152,176,220,221,124,150,204,220,73,228,194,218,57,195,173,159,173,175,206,176,229,79,164,201,203,152,214,116,137,219,66,222,214,80,175,202,90,225,113,219,206,78,190,89,214,50,209,154,188,227,194,157,195,74,186,206,130,198,73,212,60,204,122,222,99,205,196,229,213,83,230,108,171,126,192,104,216,207,117,217,197,214,79,207,110,221,79,217,144,206,160,206,172,197,183,207,217,207,113,210,221,71,161,221,164,227,214,142,177,185,180,103,130,198,123,205,74,216,102,219,160,217,75,204,114,192,213,166,188,118,222,227,92,195,219,161,200,221,69,203,143,198,198,217,198,66,212,50,208,116,199,125,210,207,167,225,116,207,97,184,99,220,184,203,184,219,177,167,202,214,55,207,161,197,122,212,226,187,96,216,201,188,135,224,207,139,225,230,220,121,221,107,212,66,170,169,210,199,102,220,94,159,184,207,92,207,231,214,125,227,220,205,58,193,203,215,223,229,78,196,170,185,196,162,234,56,201,123,171,231,196,86,162,199,213,220,68,200,68,205,88,225,135,220,82,182,215,222,79,152,230,62,162,218,184,224,67,206,99,189,124,214,197,73,204,105,221,179,102,218,232,80,214,181,170,204,165,216,207,217,212,195,176,215,106,192,160,221,182,217,57,211,88,198,233,113,171,204,138,193,209,225,59,176,184,134,223,151,193,200,217,100,225,79,180,142,190,123,222,80,232,216,133,216,148,211,110,198,96,187,224,95,208,112,178,227,94,171,96,181,209,170,225,196,206,94,216,87,217,171,191,82,218,127,227,176,219,207,230,79,214,203,105,213,143,174,188,125,193,220,60,215,172,214,101,211,110,161,117,187,180,125,218,220,62,208,203,217,87,198,156,216,226,161,161,223,224,72,178,198,213,195,219,208,140,175,217,74,201,201,66,186,154,229,89,226,169,204,87,184,85,161,133,201,80,176,188,114,224,77,207,126,202,83,219,200,125,172,169,190,216,80,88,221,68,218,133,216,117,217,157,217,170,190,124,214,210,156,231,84,207,204,113,200,70,222,162,208,227,92,223,136,167,195,221,221,77,173,213,109,214,117,211,217,89,217,91,210,152,194,206,202,110,216,177,190,207,227,185,172,230,172,207,171,199,234,207,149,194,192,179,212,209,210,101,198,225,85,164,211,110,194,182,211,224,65,228,218,79,224,81,122,208,154,129,206,92,193,171,148,188,221,80,220,161,165,166,161,214,99,210,64,174,224,221,105,200,122,230,216,94,223,128,225,161,219,126,187,137,191,222,214,148,151,198,218,210,110,208,228,184,211,35,202,218,195,216,115,212,95,177,199,101,184,208,202,212,134,193,129,192,81,182,223,70,226,230,134,167,183,198,222,227,227,226,63,213,109,187,177,219,223,203,144,179,209,103,177,181,158,221,90,222,166,207,175,230,207,99,205,234,210,210,168,223,143,210,187,209,204,150,209,213,208,193,221,214,77,215,199,81,197,82,177,190,210,231,79,179,221,64,182,199,82,204,204,95,172,187,178,209,86,222,220,118,192,223,88,220,77,174,104,224,137,182,186,96,207,198,74,152,196,217,206,79,214,208,204,180,94,215,81,177,160,201,164,173,205,76,199,220,228,91,215,155,226,79,133,181,136,182,226,96,221,109,209,223,71,202,95,217,87,202,204,183,210,187,212,81,226,184,224,88,170,214,198,226,142,212,81,209,189,172,192,221,216,123,221,126,204,218,222,76,205,73,225,221,73,204,108,201,88,174,197,136,223,90,189,56,207,147,206,212,73,201,83,204,112,137,227,67,208,137,219,225,65,200,186,99,214,97,215,74,203,65,199,216,108,216,80,206,219,104,226,180,225,199,186,197,226,157,102,177,107,231,156,141,226,70,220,216,223,64,214,66,201,174,170,207,46,202,131,173,218,125,217,157,234,192,159,174,209,95,196,224,59,220,69,211,130,203,222,88,208,86,198,127,219,228,75,218,170,168,198,128,215,54,211,167,186,117,211,162,221,219,105,223,99,223,127,202,218,213,143,194,181,200,180,230,224,97,181,132,173,202,221,57,151,220,77,220,160,206,188,101,197,72,213,95,193,212,189,105,226,100,205,201,56,211,93,178,212,88,208,83,213,165,219,183,236,121,220,210,94,212,171,186,218,137,212,129,175,203,223,134,194,95,193,191,105,229,208,102,196,120,191,221,217,65,206,200,74,168,180,199,217,119,223,68,211,125,204,105,180,164,215,227,128,211,166,218,86,185,74,214,57,200,171,111,185,73,199,220,213,192,216,107,211,115,219,227,192,221,101,203,65,211,51,216,84,193,121,214,86,195,115,179,229,90,215,92,207,63,179,212,38,202,104,182,125,179,99,147,184,210,166,227,232,164,120,218,169,203,154,192,224,217,122,160,205,206,221,80,191,217,166,202,78,206,147,202,155,195,76,204,136,191,112,195,160,147,226,91,224,216,212,177,188,165,174,130,203,221,220,133,209,147,216,69,159,155,143,213,94,227,139,209,163,183,199,112,217,213,98,217,96,185,158,173,229,51,209,195,227,214,161,213,83,168,229,209,118,221,224,59,179,161,220,209,193,199,199,212,107,226,219,204,117,166,223,122,166,181,163,176,223,176,130,223,221,202,89,188,147,160,143,218,223,206,151,201,161,130,176,175,138,126,209,112,230,94,211,17,103,218,73,218,131,210,104,214,63,222,38,135,140,215,143,215,191,185,223,207,215,203,46,219,207,93,177,85,213,191,223,56,181,209,82,210,221,66,210,195,223,184,138,217,48,194,73,150,199,220,183,209,60,194,103,218,103,211,216,124,197,217,185,106,185,207,174,165,204,138,220,68,218,151,202,68,214,155,183,221,66,216,61,218,122,214,178,202,178,217,142,215,126,187,148,219,98,180,222,217,80,210,203,43,208,154,220,101,167,206,211,212,208,72,147,225,139,174,207,36,200,234,205,211,180,205,202,126,159,186,116,211,154,192,155,194,168,198,160,218,220,202,153,222,215,66,174,128,211,104,136,171,235,219,112,156,209,109,203,132,192,181,215,112,205,68,215,82,213,117,189,221,186,211,171,208,136,189,128,210,96,199,107,195,232,74,223,132,193,198,46,220,73,181,112,224,133,221,144,224,83,232,217,131,186,53,214,225,95,203,70,102,217,106,224,79,210,113,177,150,228,220,102,225,80,221,170,206,105,223,112,210,46,201,89,197,207,128,235,111,212,161,144,221,182,200,77,213,229,90,134,223,179,212,204,125,197,215,80,233,218,44,226,53,152,184,220,113,219,216,110,214,206,151,215,224,216,163,144,190,133,223,195,216,203,67,95,169,191,131,208,78,104,176,179,148,207,172,220,98,202,118,218,204,120,213,92,213,93,210,203,219,75,212,227,212,188,187,201,100,206,151,200,96,197,215,157,210,70,207,182,205,205,101,212,117,230,86,163,143,167,189,215,168,216,194,98,218,128,219,94,149,188,217,48,172,174,131,131,182,171,200,115,220,217,91,200,80,178,188,226,49,192,205,222,127,194,134,175,214,115,212,214,82,193,106,217,48,197,114,204,114,201,221,190,174,214,168,223,86,217,212,214,181,204,96,190,189,216,91,204,118,226,110,198,224,158,195,117,189,200,150,231,206,91,209,207,118,223,183,232,146,158,207,213,122,204,118,200,103,200,227,76,179,195,73,215,93,214,170,215,232,41,210,107,138,202,204,125,198,134,225,80,117,164,185,197,106,232,74,139,216,207,209,57,207,94,200,229,190,192,140,112,208,155,191,177,216,203,142,192,103,195,219,91,179,228,187,115,213,217,192,193,215,141,218,70,186,37,225,190,84,178,177,162,218,210,185,176,195,97,218,63,218,176,227,215,149,224,221,115,208,214,133,182,188,205,163,207,58,199,217,129,208,184,214,206,133,228,200,80,224,179,229,152,208,95,194,170,224,178,196,181,94,199,90,203,75,216,153,169,213,86,225,67,192,194,65,189,201,103,188,153,163,213,226,142,185,133,226,106,181,215,199,209,211,56,204,114,181,163,171,228,228,73,74,198,203,186,178,185,125,229,221,204,41,165,189,126,205,173,116,179,198,159,216,129,209,222,174,183,229,95,212,68,177,152,217,138,156,135,105,204,193,188,127,209,150,133,163,208,80,218,232,211,77,230,37,198,224,165,213,81,220,207,195,173,174,212,102,206,117,196,178,222,134,205,216,203,86,151,184,157,217,222,123,213,159,195,121,207,159,220,120,212,176,144,217,67,173,216,105,206,90,220,121,217,59,184,219,156,213,149,143,216,221,140,225,182,115,209,115,209,45,223,172,227,133,187,165,199,168,224,91,206,115,153,202,197,62,169,210,134,215,167,203,214,141,200,213,182,90,214,170,206,199,219,54,167,154,72,194,122,181,197,129,214,105,153,209,137,202,227,72,207,61,178,127,181,210,200,46,188,210,214,67,189,216,51,209,125,190,127,208,110,191,219,137,213,76,206,120,186,121,201,222,113,195,194,68,183,179,184,223,61,180,220,197,133,208,226,136,217,200,93,178,220,113,197,198,172,141,225,102,159,149,213,196,100,220,196,176,232,182,187,171,165,182,101,175,169,191,224,110,200,128,200,129,114,179,188,165,198,216,184,174,216,67,229,198,220,32,232,219,72,219,203,127,88,212,81,142,223,210,166,97,145,209,77,216,227,196,83,202,137,214,82,223,114,205,177,183,196]
ivf.pid.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:4280414a1095fa62b511fe43062ba1090c4d82a74758bfef28ded339bad9295c
3
- size 2217688
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4ff4926e9bdd91e35198628f83dc09b888ed0aba6c736a69c14f13f4c78af455
3
+ size 2211352
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 4877 elements starting with...",
41
  [
42
  "Deep neural networks have demonstrated remarkable performance in supervised learning tasks but require large amounts of labeled data. Self-supervised learning offers an alternative paradigm, enabling the model to learn from data without explicit labels. Information theory has been instrumental in understanding and optimizing deep neural networks. Specifically, the information bottleneck principle has been applied to optimize the trade-off between compression and relevant information preservation in supervised settings. However, the optimal information objective in self-supervised learning remains unclear. In this paper, we review various approaches to self-supervised learning from an information-theoretic standpoint and present a unified framework that formalizes the self-supervised information-theoretic learning problem. We integrate existing research into a coherent framework, examine recent self-supervised methods, and identify research opportunities and challenges. Moreover, we discuss empirical measurement of information-theoretic quantities and their estimators. This paper offers a comprehensive review of the intersection between information theory, self-supervised learning, and deep neural networks.",
43
  "Pre-trained large language models (LLMs) capture procedural knowledge about the world. Recent work has leveraged LLM's ability to generate abstract plans to simplify challenging control tasks, either by action scoring, or action modeling (fine-tuning). However, the transformer architecture inherits several constraints that make it difficult for the LLM to directly serve as the agent: e.g. limited input lengths, fine-tuning inefficiency, bias from pre-training, and incompatibility with non-text environments. To maintain compatibility with a low-level trainable actor, we propose to instead use the knowledge in LLMs to simplify the control problem, rather than solving it. We propose the Plan, Eliminate, and Track (PET) framework. The Plan module translates a task description into a list of high-level sub-tasks. The Eliminate module masks out irrelevant objects and receptacles from the observation for the current sub-task. Finally, the Track module determines whether the agent has accomplished each sub-task. On the AlfWorld instruction following benchmark, the PET framework leads to a significant 15% improvement over SOTA for generalization to human goal specifications.",
@@ -50,7 +50,7 @@
50
  "root":".ragatouille/",
51
  "experiment":"colbert",
52
  "index_root":null,
53
- "name":"2024-08/23/11.54.44",
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":835164,
63
- "avg_doclen":171.2454377691,
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 4878 elements starting with...",
41
  [
42
  "Deep neural networks have demonstrated remarkable performance in supervised learning tasks but require large amounts of labeled data. Self-supervised learning offers an alternative paradigm, enabling the model to learn from data without explicit labels. Information theory has been instrumental in understanding and optimizing deep neural networks. Specifically, the information bottleneck principle has been applied to optimize the trade-off between compression and relevant information preservation in supervised settings. However, the optimal information objective in self-supervised learning remains unclear. In this paper, we review various approaches to self-supervised learning from an information-theoretic standpoint and present a unified framework that formalizes the self-supervised information-theoretic learning problem. We integrate existing research into a coherent framework, examine recent self-supervised methods, and identify research opportunities and challenges. Moreover, we discuss empirical measurement of information-theoretic quantities and their estimators. This paper offers a comprehensive review of the intersection between information theory, self-supervised learning, and deep neural networks.",
43
  "Pre-trained large language models (LLMs) capture procedural knowledge about the world. Recent work has leveraged LLM's ability to generate abstract plans to simplify challenging control tasks, either by action scoring, or action modeling (fine-tuning). However, the transformer architecture inherits several constraints that make it difficult for the LLM to directly serve as the agent: e.g. limited input lengths, fine-tuning inefficiency, bias from pre-training, and incompatibility with non-text environments. To maintain compatibility with a low-level trainable actor, we propose to instead use the knowledge in LLMs to simplify the control problem, rather than solving it. We propose the Plan, Eliminate, and Track (PET) framework. The Plan module translates a task description into a list of high-level sub-tasks. The Eliminate module masks out irrelevant objects and receptacles from the observation for the current sub-task. Finally, the Track module determines whether the agent has accomplished each sub-task. On the AlfWorld instruction following benchmark, the PET framework leads to a significant 15% improvement over SOTA for generalization to human goal specifications.",
 
50
  "root":".ragatouille/",
51
  "experiment":"colbert",
52
  "index_root":null,
53
+ "name":"2024-08/23/20.56.06",
54
  "rank":0,
55
  "nranks":1,
56
  "amp":true,
 
59
  },
60
  "num_chunks":1,
61
  "num_partitions":8192,
62
+ "num_embeddings":835360,
63
+ "avg_doclen":171.2505125051,
64
  "RAGatouille":{
65
  "index_config":{
66
  "index_type":"PLAID",
pid_docid_map.json CHANGED
@@ -4875,5 +4875,6 @@
4875
  "4873":"2408.09787",
4876
  "4874":"2408.12076",
4877
  "4875":"2408.12480",
4878
- "4876":"2408.12503"
 
4879
  }
 
4875
  "4873":"2408.09787",
4876
  "4874":"2408.12076",
4877
  "4875":"2408.12480",
4878
+ "4876":"2408.12503",
4879
+ "4877":"2408.12060"
4880
  }
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 4877 elements starting with...",
41
  [
42
  "Deep neural networks have demonstrated remarkable performance in supervised learning tasks but require large amounts of labeled data. Self-supervised learning offers an alternative paradigm, enabling the model to learn from data without explicit labels. Information theory has been instrumental in understanding and optimizing deep neural networks. Specifically, the information bottleneck principle has been applied to optimize the trade-off between compression and relevant information preservation in supervised settings. However, the optimal information objective in self-supervised learning remains unclear. In this paper, we review various approaches to self-supervised learning from an information-theoretic standpoint and present a unified framework that formalizes the self-supervised information-theoretic learning problem. We integrate existing research into a coherent framework, examine recent self-supervised methods, and identify research opportunities and challenges. Moreover, we discuss empirical measurement of information-theoretic quantities and their estimators. This paper offers a comprehensive review of the intersection between information theory, self-supervised learning, and deep neural networks.",
43
  "Pre-trained large language models (LLMs) capture procedural knowledge about the world. Recent work has leveraged LLM's ability to generate abstract plans to simplify challenging control tasks, either by action scoring, or action modeling (fine-tuning). However, the transformer architecture inherits several constraints that make it difficult for the LLM to directly serve as the agent: e.g. limited input lengths, fine-tuning inefficiency, bias from pre-training, and incompatibility with non-text environments. To maintain compatibility with a low-level trainable actor, we propose to instead use the knowledge in LLMs to simplify the control problem, rather than solving it. We propose the Plan, Eliminate, and Track (PET) framework. The Plan module translates a task description into a list of high-level sub-tasks. The Eliminate module masks out irrelevant objects and receptacles from the observation for the current sub-task. Finally, the Track module determines whether the agent has accomplished each sub-task. On the AlfWorld instruction following benchmark, the PET framework leads to a significant 15% improvement over SOTA for generalization to human goal specifications.",
@@ -50,7 +50,7 @@
50
  "root": ".ragatouille\/",
51
  "experiment": "colbert",
52
  "index_root": null,
53
- "name": "2024-08\/23\/11.54.44",
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": 835163.9992828369,
63
- "avg_doclen_est": 171.2454376220703
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 4878 elements starting with...",
41
  [
42
  "Deep neural networks have demonstrated remarkable performance in supervised learning tasks but require large amounts of labeled data. Self-supervised learning offers an alternative paradigm, enabling the model to learn from data without explicit labels. Information theory has been instrumental in understanding and optimizing deep neural networks. Specifically, the information bottleneck principle has been applied to optimize the trade-off between compression and relevant information preservation in supervised settings. However, the optimal information objective in self-supervised learning remains unclear. In this paper, we review various approaches to self-supervised learning from an information-theoretic standpoint and present a unified framework that formalizes the self-supervised information-theoretic learning problem. We integrate existing research into a coherent framework, examine recent self-supervised methods, and identify research opportunities and challenges. Moreover, we discuss empirical measurement of information-theoretic quantities and their estimators. This paper offers a comprehensive review of the intersection between information theory, self-supervised learning, and deep neural networks.",
43
  "Pre-trained large language models (LLMs) capture procedural knowledge about the world. Recent work has leveraged LLM's ability to generate abstract plans to simplify challenging control tasks, either by action scoring, or action modeling (fine-tuning). However, the transformer architecture inherits several constraints that make it difficult for the LLM to directly serve as the agent: e.g. limited input lengths, fine-tuning inefficiency, bias from pre-training, and incompatibility with non-text environments. To maintain compatibility with a low-level trainable actor, we propose to instead use the knowledge in LLMs to simplify the control problem, rather than solving it. We propose the Plan, Eliminate, and Track (PET) framework. The Plan module translates a task description into a list of high-level sub-tasks. The Eliminate module masks out irrelevant objects and receptacles from the observation for the current sub-task. Finally, the Track module determines whether the agent has accomplished each sub-task. On the AlfWorld instruction following benchmark, the PET framework leads to a significant 15% improvement over SOTA for generalization to human goal specifications.",
 
50
  "root": ".ragatouille\/",
51
  "experiment": "colbert",
52
  "index_root": null,
53
+ "name": "2024-08\/23\/20.56.06",
54
  "rank": 0,
55
  "nranks": 1,
56
  "amp": true,
 
59
  },
60
  "num_chunks": 1,
61
  "num_partitions": 8192,
62
+ "num_embeddings_est": 835360.0307006836,
63
+ "avg_doclen_est": 171.25051879882812
64
  }