Evolving Deeper LLM Thinking
Paper
•
2501.09891
•
Published
•
106
LLM
-
-
PaSa: An LLM Agent for Comprehensive Academic Paper Search
Paper • 2501.10120 • Published • 43 -
Multiple Choice Questions: Reasoning Makes Large Language Models (LLMs) More Self-Confident Even When They Are Wrong
Paper • 2501.09775 • Published • 29 -
ComplexFuncBench: Exploring Multi-Step and Constrained Function Calling under Long-Context Scenario
Paper • 2501.10132 • Published • 19 -
HiFi-SR: A Unified Generative Transformer-Convolutional Adversarial Network for High-Fidelity Speech Super-Resolution
Paper • 2501.10045 • Published • 9 -
X-Dyna: Expressive Dynamic Human Image Animation
Paper • 2501.10021 • Published • 14 -
GameFactory: Creating New Games with Generative Interactive Videos
Paper • 2501.08325 • Published • 64 -
VideoWorld: Exploring Knowledge Learning from Unlabeled Videos
Paper • 2501.09781 • Published • 25 -
The Lessons of Developing Process Reward Models in Mathematical Reasoning
Paper • 2501.07301 • Published • 91 -
Do generative video models learn physical principles from watching videos?
Paper • 2501.09038 • Published • 32 -
Towards Large Reasoning Models: A Survey of Reinforced Reasoning with Large Language Models
Paper • 2501.09686 • Published • 36 -
Inference-Time Scaling for Diffusion Models beyond Scaling Denoising Steps
Paper • 2501.09732 • Published • 68 -
FAST: Efficient Action Tokenization for Vision-Language-Action Models
Paper • 2501.09747 • Published • 23 -
OmniThink: Expanding Knowledge Boundaries in Machine Writing through Thinking
Paper • 2501.09751 • Published • 47 -
Agent-R: Training Language Model Agents to Reflect via Iterative Self-Training
Paper • 2501.11425 • Published • 91 -
TokenVerse: Versatile Multi-concept Personalization in Token Modulation Space
Paper • 2501.12224 • Published • 46 -
InternLM-XComposer2.5-Reward: A Simple Yet Effective Multi-Modal Reward Model
Paper • 2501.12368 • Published • 42 -
Hunyuan3D 2.0: Scaling Diffusion Models for High Resolution Textured 3D Assets Generation
Paper • 2501.12202 • Published • 33 -
Reasoning Language Models: A Blueprint
Paper • 2501.11223 • Published • 32 -
Mobile-Agent-E: Self-Evolving Mobile Assistant for Complex Tasks
Paper • 2501.11733 • Published • 28 -
Learn-by-interact: A Data-Centric Framework for Self-Adaptive Agents in Realistic Environments
Paper • 2501.10893 • Published • 24 -
Go-with-the-Flow: Motion-Controllable Video Diffusion Models Using Real-Time Warped Noise
Paper • 2501.08331 • Published • 20 -
Taming Teacher Forcing for Masked Autoregressive Video Generation
Paper • 2501.12389 • Published • 10 -
The Geometry of Tokens in Internal Representations of Large Language Models
Paper • 2501.10573 • Published • 9 -
Fixing Imbalanced Attention to Mitigate In-Context Hallucination of Large Vision-Language Model
Paper • 2501.12206 • Published • 4 -
Test-Time Preference Optimization: On-the-Fly Alignment via Iterative Textual Feedback
Paper • 2501.12895 • Published • 56 -
Autonomy-of-Experts Models
Paper • 2501.13074 • Published • 41 -
Pairwise RM: Perform Best-of-N Sampling with Knockout Tournament
Paper • 2501.13007 • Published • 20 -
O1-Pruner: Length-Harmonizing Fine-Tuning for O1-Like Reasoning Pruning
Paper • 2501.12570 • Published • 24 -
Improving Video Generation with Human Feedback
Paper • 2501.13918 • Published • 49 -
Temporal Preference Optimization for Long-Form Video Understanding
Paper • 2501.13919 • Published • 22 -
Step-KTO: Optimizing Mathematical Reasoning through Stepwise Binary Feedback
Paper • 2501.10799 • Published • 15 -
Control LLM: Controlled Evolution for Intelligence Retention in LLM
Paper • 2501.10979 • Published • 6 -
SRMT: Shared Memory for Multi-agent Lifelong Pathfinding
Paper • 2501.13200 • Published • 63 -
RL + Transformer = A General-Purpose Problem Solver
Paper • 2501.14176 • Published • 24 -
RealCritic: Towards Effectiveness-Driven Evaluation of Language Model Critiques
Paper • 2501.14492 • Published • 30 -
Chain-of-Retrieval Augmented Generation
Paper • 2501.14342 • Published • 51 -
Humanity's Last Exam
Paper • 2501.14249 • Published • 62 -
Towards General-Purpose Model-Free Reinforcement Learning
Paper • 2501.16142 • Published • 26 -
ARWKV: Pretrain is not what we need, an RNN-Attention-Based Language Model Born from Transformer
Paper • 2501.15570 • Published • 23 -
Visual Generation Without Guidance
Paper • 2501.15420 • Published • 8 -
Feasible Learning
Paper • 2501.14912 • Published • 5 -
Return of the Encoder: Maximizing Parameter Efficiency for SLMs
Paper • 2501.16273 • Published • 5 -
Large Concept Models: Language Modeling in a Sentence Representation Space
Paper • 2412.08821 • Published • 14 -
Over-Tokenized Transformer: Vocabulary is Generally Worth Scaling
Paper • 2501.16975 • Published • 26 -
Open Problems in Mechanistic Interpretability
Paper • 2501.16496 • Published • 19 -
SFT Memorizes, RL Generalizes: A Comparative Study of Foundation Model Post-training
Paper • 2501.17161 • Published • 106 -
Critique Fine-Tuning: Learning to Critique is More Effective than Learning to Imitate
Paper • 2501.17703 • Published • 55 -
Trading Inference-Time Compute for Adversarial Robustness
Paper • 2501.18841 • Published • 3 -
SAeUron: Interpretable Concept Unlearning in Diffusion Models with Sparse Autoencoders
Paper • 2501.18052 • Published • 6 -
The Surprising Agreement Between Convex Optimization Theory and Learning-Rate Scheduling for Large Model Training
Paper • 2501.18965 • Published • 6 -
CowPilot: A Framework for Autonomous and Human-Agent Collaborative Web Navigation
Paper • 2501.16609 • Published • 6 -
s1: Simple test-time scaling
Paper • 2501.19393 • Published • 105 -
Reward-Guided Speculative Decoding for Efficient LLM Reasoning
Paper • 2501.19324 • Published • 37 -
PixelWorld: Towards Perceiving Everything as Pixels
Paper • 2501.19339 • Published • 16 -
Self-supervised Quantized Representation for Seamlessly Integrating Knowledge Graphs with Large Language Models
Paper • 2501.18119 • Published • 24 -
DINO-WM: World Models on Pre-trained Visual Features enable Zero-shot Planning
Paper • 2411.04983 • Published • 11 -
Constitutional Classifiers: Defending against Universal Jailbreaks across Thousands of Hours of Red Teaming
Paper • 2501.18837 • Published • 9 -
o3-mini vs DeepSeek-R1: Which One is Safer?
Paper • 2501.18438 • Published • 22 -
SANA 1.5: Efficient Scaling of Training-Time and Inference-Time Compute in Linear Diffusion Transformer
Paper • 2501.18427 • Published • 16 -
WILDCHAT-50M: A Deep Dive Into the Role of Synthetic Data in Post-Training
Paper • 2501.18511 • Published • 19 -
Streaming DiLoCo with overlapping communication: Towards a Distributed Free Lunch
Paper • 2501.18512 • Published • 27 -
Large Language Models Think Too Fast To Explore Effectively
Paper • 2501.18009 • Published • 23 -
GuardReasoner: Towards Reasoning-based LLM Safeguards
Paper • 2501.18492 • Published • 81 -
Thoughts Are All Over the Place: On the Underthinking of o1-Like LLMs
Paper • 2501.18585 • Published • 55 -
Preference Leakage: A Contamination Problem in LLM-as-a-judge
Paper • 2502.01534 • Published • 37 -
AlignVLM: Bridging Vision and Language Latent Spaces for Multimodal Understanding
Paper • 2502.01341 • Published • 35 -
MakeAnything: Harnessing Diffusion Transformers for Multi-Domain Procedural Sequence Generation
Paper • 2502.01572 • Published • 20 -
DeepRAG: Thinking to Retrieval Step by Step for Large Language Models
Paper • 2502.01142 • Published • 23 -
ZebraLogic: On the Scaling Limits of LLMs for Logical Reasoning
Paper • 2502.01100 • Published • 15 -
The Jumping Reasoning Curve? Tracking the Evolution of Reasoning Performance in GPT-[n] and o-[n] Models on Multimodal Puzzles
Paper • 2502.01081 • Published • 14 -
PhD Knowledge Not Required: A Reasoning Challenge for Large Language Models
Paper • 2502.01584 • Published • 9 -
Improving Transformer World Models for Data-Efficient RL
Paper • 2502.01591 • Published • 9 -
Improved Training Technique for Latent Consistency Models
Paper • 2502.01441 • Published • 7 -
Lifelong Sequential Knowledge Editing without Model Degradation
Paper • 2502.01636 • Published • 5 -
Language Models Prefer What They Know: Relative Confidence Estimation via Confidence Preferences
Paper • 2502.01126 • Published • 4 -
LongDPO: Unlock Better Long-form Generation Abilities for LLMs via Critique-augmented Stepwise Information
Paper • 2502.02095 • Published • 4 -
The Differences Between Direct Alignment Algorithms are a Blur
Paper • 2502.01237 • Published • 111 -
Process Reinforcement through Implicit Rewards
Paper • 2502.01456 • Published • 54 -
ACECODER: Acing Coder RL via Automated Test-Case Synthesis
Paper • 2502.01718 • Published • 28 -
Satori: Reinforcement Learning with Chain-of-Action-Thought Enhances LLM Reasoning via Autoregressive Search
Paper • 2502.02508 • Published • 21 -
QLASS: Boosting Language Agent Inference via Q-Guided Stepwise Search
Paper • 2502.02584 • Published • 16 -
COCONut-PanCap: Joint Panoptic Segmentation and Grounded Captions for Fine-Grained Understanding and Generation
Paper • 2502.02589 • Published • 9 -
Sample, Scrutinize and Scale: Effective Inference-Time Search by Scaling Verification
Paper • 2502.01839 • Published • 5 -
LIMO: Less is More for Reasoning
Paper • 2502.03387 • Published • 56 -
Demystifying Long Chain-of-Thought Reasoning in LLMs
Paper • 2502.03373 • Published • 51 -
SmolLM2: When Smol Goes Big -- Data-Centric Training of a Small Language Model
Paper • 2502.02737 • Published • 187 -
Boosting Multimodal Reasoning with MCTS-Automated Structured Thinking
Paper • 2502.02339 • Published • 22 -
On Teacher Hacking in Language Model Distillation
Paper • 2502.02671 • Published • 17 -
Token Assorted: Mixing Latent and Text Tokens for Improved Language Model Reasoning
Paper • 2502.03275 • Published • 13 -
A Probabilistic Inference Approach to Inference-Time Scaling of LLMs using Particle-Based Monte Carlo Methods
Paper • 2502.01618 • Published • 9 -
Jailbreaking with Universal Multi-Prompts
Paper • 2502.01154 • Published • 8 -
ConceptAttention: Diffusion Transformers Learn Highly Interpretable Features
Paper • 2502.04320 • Published • 33 -
Gold-medalist Performance in Solving Olympiad Geometry with AlphaGeometry2
Paper • 2502.03544 • Published • 42 -
Analyze Feature Flow to Enhance Interpretation and Steering in Language Models
Paper • 2502.03032 • Published • 55 -
Llasa: Scaling Train-Time and Inference-Time Compute for Llama-based Speech Synthesis
Paper • 2502.04128 • Published • 22 -
BOLT: Bootstrap Long Chain-of-Thought in Language Models without Distillation
Paper • 2502.03860 • Published • 22 -
Ola: Pushing the Frontiers of Omni-Modal Language Model with Progressive Modality Alignment
Paper • 2502.04328 • Published • 25 -
Weak-to-Strong Diffusion with Reflection
Paper • 2502.00473 • Published • 20 -
ScoreFlow: Mastering LLM Agent Workflows via Score-based Preference Optimization
Paper • 2502.04306 • Published • 17 -
UltraIF: Advancing Instruction Following from the Wild
Paper • 2502.04153 • Published • 20 -
PILAF: Optimal Human Preference Sampling for Reward Modeling
Paper • 2502.04270 • Published • 11 -
Beyond Prompt Content: Enhancing LLM Performance via Content-Format Integrated Prompt Optimization
Paper • 2502.04295 • Published • 11 -
Scaling up Test-Time Compute with Latent Reasoning: A Recurrent Depth Approach
Paper • 2502.05171 • Published • 112 -
Goku: Flow Based Video Generative Foundation Models
Paper • 2502.04896 • Published • 85 -
VideoRoPE: What Makes for Good Video Rotary Position Embedding?
Paper • 2502.05173 • Published • 60 -
Step Back to Leap Forward: Self-Backtracking for Boosting Reasoning of Language Models
Paper • 2502.04404 • Published • 19 -
Generating Symbolic World Models via Test-time Scaling of Large Language Models
Paper • 2502.04728 • Published • 16 -
ARR: Question Answering with Large Language Models via Analyzing, Retrieving, and Reasoning
Paper • 2502.04689 • Published • 7 -
Value-Based Deep RL Scales Predictably
Paper • 2502.04327 • Published • 5 -
YINYANG-ALIGN: Benchmarking Contradictory Objectives and Proposing Multi-Objective Optimization based DPO for Text-to-Image Alignment
Paper • 2502.03512 • Published • 5 -
SPARC: Subspace-Aware Prompt Adaptation for Robust Continual Learning in LLMs
Paper • 2502.02909 • Published • 2 -
Competitive Programming with Large Reasoning Models
Paper • 2502.06807 • Published • 59 -
LLMs Can Easily Learn to Reason from Demonstrations Structure, not content, is what matters!
Paper • 2502.07374 • Published • 31 -
Teaching Language Models to Critique via Reinforcement Learning
Paper • 2502.03492 • Published • 22 -
Mask-Enhanced Autoregressive Prediction: Pay Less Attention to Learn More
Paper • 2502.07490 • Published • 9 -
The Hidden Life of Tokens: Reducing Hallucination of Large Vision-Language Models via Visual Information Steering
Paper • 2502.03628 • Published • 11 -
History-Guided Video Diffusion
Paper • 2502.06764 • Published • 10 -
Can 1B LLM Surpass 405B LLM? Rethinking Compute-Optimal Test-Time Scaling
Paper • 2502.06703 • Published • 132 -
Exploring the Limit of Outcome Reward for Learning Mathematical Reasoning
Paper • 2502.06781 • Published • 56 -
LM2: Large Memory Models
Paper • 2502.06049 • Published • 26 -
The Curse of Depth in Large Language Models
Paper • 2502.05795 • Published • 29 -
ReasonFlux: Hierarchical LLM Reasoning via Scaling Thought Templates
Paper • 2502.06772 • Published • 18 -
Distillation Scaling Laws
Paper • 2502.08606 • Published • 40 -
LLM Pretraining with Continuous Concepts
Paper • 2502.08524 • Published • 24 -
Ignore the KL Penalty! Boosting Exploration on Critical Tokens to Enhance RL Fine-Tuning
Paper • 2502.06533 • Published • 17 -
DPO-Shift: Shifting the Distribution of Direct Preference Optimization
Paper • 2502.07599 • Published • 14 -
Mediator: Memory-efficient LLM Merging with Less Parameter Conflicts and Uncertainty Based Routing
Paper • 2502.04411 • Published • 4 -
Towards Trustworthy Retrieval Augmented Generation for Large Language Models: A Survey
Paper • 2502.06872 • Published • 8 -
Logical Reasoning in Large Language Models: A Survey
Paper • 2502.09100 • Published • 20 -
CoT-Valve: Length-Compressible Chain-of-Thought Tuning
Paper • 2502.09601 • Published • 11 -
Mathematical Reasoning in Large Language Models: Assessing Logical and Arithmetic Errors across Wide Numerical Ranges
Paper • 2502.08680 • Published • 10 -
MME-CoT: Benchmarking Chain-of-Thought in Large Multimodal Models for Reasoning Quality, Robustness, and Efficiency
Paper • 2502.09621 • Published • 26 -
Large Language Diffusion Models
Paper • 2502.09992 • Published • 74 -
The Danger of Overthinking: Examining the Reasoning-Action Dilemma in Agentic Tasks
Paper • 2502.08235 • Published • 50 -
MM-RLHF: The Next Step Forward in Multimodal LLM Alignment
Paper • 2502.10391 • Published • 29 -
Diverse Inference and Verification for Advanced Reasoning
Paper • 2502.09955 • Published • 16 -
Precise Parameter Localization for Textual Generation in Diffusion Models
Paper • 2502.09935 • Published • 11 -
We Can't Understand AI Using our Existing Vocabulary
Paper • 2502.07586 • Published • 8 -
CRANE: Reasoning with constrained LLM generation
Paper • 2502.09061 • Published • 18 -
Dyve: Thinking Fast and Slow for Dynamic Process Verification
Paper • 2502.11157 • Published • 6 -
One Example Shown, Many Concepts Known! Counterexample-Driven Conceptual Reasoning in Mathematical LLMs
Paper • 2502.10454 • Published • 6 -
Diffusion Models without Classifier-free Guidance
Paper • 2502.12154 • Published • 3 -
Large Language Models and Mathematical Reasoning Failures
Paper • 2502.11574 • Published • 3 -
Continuous Diffusion Model for Language Modeling
Paper • 2502.11564 • Published • 47 -
Revisiting the Test-Time Scaling of o1-like Models: Do they Truly Possess Test-Time Scaling Capabilities?
Paper • 2502.12215 • Published • 11 -
Training Language Models to Reason Efficiently
Paper • 2502.04463 • Published -
Efficient Reasoning with Hidden Thinking
Paper • 2501.19201 • Published