new

Get trending papers in your email inbox once a day!

Get trending papers in your email inbox!

Daily Papers

by AK and the research community

Feb 18

Submitted by

HelloJiang

Native Sparse Attention: Hardware-Aligned and Natively Trainable Sparse Attention

·
15 authors

Submitted by

akhaliq

SWE-Lancer: Can Frontier LLMs Earn $1 Million from Real-World Freelance Software Engineering?

·
4 authors

Submitted by

RunpeiDong

Learning Getting-Up Policies for Real-World Humanoid Robots

·
4 authors

Submitted by

Ningyu

ReLearn: Unlearning via Learning for Large Language Models

·
10 authors

Submitted by

Mifucius

I Think, Therefore I Diffuse: Enabling Multimodal In-Context Reasoning in Diffusion Models

·
8 authors

Submitted by

Ningyu

How Do LLMs Acquire New Knowledge? A Knowledge Circuits Perspective on Continual Pre-Training

·
8 authors

Submitted by

zhihz0535

IHEval: Evaluating Language Models on Following the Instruction Hierarchy

·
14 authors

Submitted by

tarsur909

CRANE: Reasoning with constrained LLM generation

·
5 authors

Submitted by

comin

HermesFlow: Seamlessly Closing the Gap in Multimodal Understanding and Generation

·
7 authors

Submitted by

Minbyul

System Message Generation for User Preferences using Open-Source Models

·
5 authors

Submitted by

aboots

Ask in Any Modality: A Comprehensive Survey on Multimodal Retrieval-Augmented Generation

·
8 authors

Submitted by

comin

Diffusion-Sharpening: Fine-tuning Diffusion Models with Denoising Trajectory Sharpening

·
6 authors

Submitted by

nielsr

Intuitive physics understanding emerges from self-supervised pretraining on natural videos

·
8 authors

Submitted by

Bohan22

SURGE: On the Potential of Large Language Models as General-Purpose Surrogate Code Executors

·
3 authors

Submitted by

WenDingY

The Mirage of Model Editing: Revisiting Evaluation in the Wild

·
8 authors

Submitted by

akhaliq

Talk Structurally, Act Hierarchically: A Collaborative Framework for LLM Multi-Agent Systems

·
5 authors

Submitted by

dreamerdeo

Sailor2: Sailing in South-East Asia with Inclusive Multilingual LLMs

·
41 authors

Submitted by

vardaan123

Explorer: Scaling Exploration-driven Web Trajectory Synthesis for Multimodal Web Agents

·
8 authors

Submitted by

akhaliq

video-SALMONN-o1: Reasoning-enhanced Audio-visual Large Language Model

·
8 authors

Submitted by

chaoyue7

MagicArticulate: Make Your 3D Models Articulation-Ready

·
11 authors

Submitted by

ingeol

SAFE-SQL: Self-Augmented In-Context Learning with Fine-grained Example Selection for Text-to-SQL

·
4 authors

Submitted by

Jianyuan1

Dyve: Thinking Fast and Slow for Dynamic Process Verification

·
5 authors

Submitted by

akhaliq

One Example Shown, Many Concepts Known! Counterexample-Driven Conceptual Reasoning in Mathematical LLMs

·
13 authors

Submitted by

KomeijiForce

Cuckoo: An IE Free Rider Hatched by Massive Nutrition in LLM's Nest

·
4 authors

Submitted by

shizhuo2

Building A Proof-Oriented Programmer That Is 64% Better Than GPT-4o Under Data Scarsity

·
3 authors

Submitted by

gkakogeorgiou

EQ-VAE: Equivariance Regularized Latent Space for Improved Generative Image Modeling

·
4 authors

Submitted by

avanturist

Memory, Benchmark & Robots: A Benchmark for Solving Complex Tasks with Reinforcement Learning

·
4 authors

Submitted by

ChengyouJia

PhysReason: A Comprehensive Benchmark towards Physics-Based Reasoning

·
9 authors

Submitted by

stojnvla

ILIAS: Instance-Level Image retrieval At Scale

·
10 authors

Submitted by

emrecanacikgoz

Can a Single Model Master Both Multi-turn Conversations and Tool Use? CALM: A Unified Conversational Agentic Language Model

·
9 authors

Submitted by

gretawarren

Show Me the Work: Fact-Checkers' Requirements for Explainable Automated Fact-Checking

·
3 authors

Submitted by

tzco

Diffusion Models without Classifier-free Guidance

·
4 authors

Submitted by

hammh0a

Towards Data-Efficient Pretraining for Atomic Property Prediction

·
3 authors

Submitted by

birgermoell

Large Language Models and Mathematical Reasoning Failures

·
2 authors

Submitted by

ishikaa

Data Valuation using Neural Networks for Efficient Instruction Fine-Tuning

·
2 authors

Submitted by

flxst

Better Embeddings with Coupled Adam

·
2 authors

Submitted by

ryuryukke

ExaGPT: Example-Based Machine-Generated Text Detection for Human Interpretability

·
5 authors

Submitted by

birgermoell

Language Complexity Measurement as a Noisy Zero-Shot Proxy for Evaluating LLM Performance

·
2 authors