Training Software Engineering Agents and Verifiers with SWE-Gym
Paper
•
2412.21139
•
Published
•
24
Synthetic Data and Self-Improvement
-
-
Evaluating Language Models as Synthetic Data Generators
Paper • 2412.03679 • Published • 48 -
Self-Rewarding Language Models
Paper • 2401.10020 • Published • 151 -
Self-Discover: Large Language Models Self-Compose Reasoning Structures
Paper • 2402.03620 • Published • 117 -
OS-Copilot: Towards Generalist Computer Agents with Self-Improvement
Paper • 2402.07456 • Published • 46 -
Learning From Mistakes Makes LLM Better Reasoner
Paper • 2310.20689 • Published • 29 -
Best Practices and Lessons Learned on Synthetic Data for Language Models
Paper • 2404.07503 • Published • 31 -
Direct Nash Optimization: Teaching Language Models to Self-Improve with General Preferences
Paper • 2404.03715 • Published • 62 -
ChatGLM-Math: Improving Math Problem-Solving in Large Language Models with a Self-Critique Pipeline
Paper • 2404.02893 • Published • 22 -
Voyager: An Open-Ended Embodied Agent with Large Language Models
Paper • 2305.16291 • Published • 11 -
Reflexion: Language Agents with Verbal Reinforcement Learning
Paper • 2303.11366 • Published • 5 -
Replacing Judges with Juries: Evaluating LLM Generations with a Panel of Diverse Models
Paper • 2404.18796 • Published • 71 -
Extending Llama-3's Context Ten-Fold Overnight
Paper • 2404.19553 • Published • 34 -
Diffusion for World Modeling: Visual Details Matter in Atari
Paper • 2405.12399 • Published • 30 -
Perplexed by Perplexity: Perplexity-Based Data Pruning With Small Reference Models
Paper • 2405.20541 • Published • 24 -
ShareGPT4Video: Improving Video Understanding and Generation with Better Captions
Paper • 2406.04325 • Published • 75 -
Depth Anything V2
Paper • 2406.09414 • Published • 103 -
Magpie: Alignment Data Synthesis from Scratch by Prompting Aligned LLMs with Nothing
Paper • 2406.08464 • Published • 71 -
Scaling Synthetic Data Creation with 1,000,000,000 Personas
Paper • 2406.20094 • Published • 104 -
Diffusion Augmented Agents: A Framework for Efficient Exploration and Transfer Learning
Paper • 2407.20798 • Published • 24 -
Mutual Reasoning Makes Smaller LLMs Stronger Problem-Solvers
Paper • 2408.06195 • Published • 73 -
Data curation via joint example selection further accelerates multimodal learning
Paper • 2406.17711 • Published • 3 -
Training Language Models to Self-Correct via Reinforcement Learning
Paper • 2409.12917 • Published • 140 -
Thinking LLMs: General Instruction Following with Thought Generation
Paper • 2410.10630 • Published • 21 -
How to Synthesize Text Data without Model Collapse?
Paper • 2412.14689 • Published • 52 -
Euclid: Supercharging Multimodal LLMs with Synthetic High-Fidelity Visual Descriptions
Paper • 2412.08737 • Published • 54 -
OS-Genesis: Automating GUI Agent Trajectory Construction via Reverse Task Synthesis
Paper • 2412.19723 • Published • 87 -
ProgCo: Program Helps Self-Correction of Large Language Models
Paper • 2501.01264 • Published • 26 -
RobustFT: Robust Supervised Fine-tuning for Large Language Models under Noisy Response
Paper • 2412.14922 • Published • 88 -
B-STaR: Monitoring and Balancing Exploration and Exploitation in Self-Taught Reasoners
Paper • 2412.17256 • Published • 47 -
Diving into Self-Evolving Training for Multimodal Reasoning
Paper • 2412.17451 • Published • 43 -
OpenAI o1 System Card
Paper • 2412.16720 • Published • 35 -
ResearchTown: Simulator of Human Research Community
Paper • 2412.17767 • Published • 14 -
SPaR: Self-Play with Tree-Search Refinement to Improve Instruction-Following in Large Language Models
Paper • 2412.11605 • Published • 18 -
AgentTrek: Agent Trajectory Synthesis via Guiding Replay with Web Tutorials
Paper • 2412.09605 • Published • 29 -
SRA-MCTS: Self-driven Reasoning Augmentation with Monte Carlo Tree Search for Code Generation
Paper • 2411.11053 • Published • 4 -
CodeDPO: Aligning Code Models with Self Generated and Verified Source Code
Paper • 2410.05605 • Published • 1 -
Enhancing LLM Reasoning via Critique Models with Test-Time and Training-Time Supervision
Paper • 2411.16579 • Published • 3 -
Vision-Language Models Can Self-Improve Reasoning via Reflection
Paper • 2411.00855 • Published • 5 -
Language Models are Hidden Reasoners: Unlocking Latent Reasoning Capabilities via Self-Rewarding
Paper • 2411.04282 • Published • 37 -
Insight-V: Exploring Long-Chain Visual Reasoning with Multimodal Large Language Models
Paper • 2411.14432 • Published • 25 -
Critic-V: VLM Critics Help Catch VLM Errors in Multimodal Reasoning
Paper • 2411.18203 • Published • 41 -
From Generation to Judgment: Opportunities and Challenges of LLM-as-a-judge
Paper • 2411.16594 • Published • 41 -
LLaMA-Berry: Pairwise Optimization for O1-like Olympiad-Level Mathematical Reasoning
Paper • 2410.02884 • Published • 54 -
Is Your LLM Secretly a World Model of the Internet? Model-Based Planning for Web Agents
Paper • 2411.06559 • Published • 16 -
Generative World Explorer
Paper • 2411.11844 • Published • 77 -
Large Language Models Can Self-Improve in Long-context Reasoning
Paper • 2411.08147 • Published • 66 -
Stronger Models are NOT Stronger Teachers for Instruction Tuning
Paper • 2411.07133 • Published • 38 -
BLIP3-KALE: Knowledge Augmented Large-Scale Dense Captions
Paper • 2411.07461 • Published • 23 -
Self-Consistency Preference Optimization
Paper • 2411.04109 • Published • 19 -
Towards System 2 Reasoning in LLMs: Learning How to Think With Meta Chain-of-Though
Paper • 2501.04682 • Published • 99 -
Enabling Scalable Oversight via Self-Evolving Critic
Paper • 2501.05727 • Published • 75 -
Multiagent Finetuning: Self Improvement with Diverse Reasoning Chains
Paper • 2501.05707 • Published • 20 -
Agent-R: Training Language Model Agents to Reflect via Iterative Self-Training
Paper • 2501.11425 • Published • 107 -
Learn-by-interact: A Data-Centric Framework for Self-Adaptive Agents in Realistic Environments
Paper • 2501.10893 • Published • 26 -
IntellAgent: A Multi-Agent Framework for Evaluating Conversational AI Systems
Paper • 2501.11067 • Published • 13 -
RealCritic: Towards Effectiveness-Driven Evaluation of Language Model Critiques
Paper • 2501.14492 • Published • 33 -
Baichuan-Omni-1.5 Technical Report
Paper • 2501.15368 • Published • 62 -
Critique Fine-Tuning: Learning to Critique is More Effective than Learning to Imitate
Paper • 2501.17703 • Published • 58 -
Atla Selene Mini: A General Purpose Evaluation Model
Paper • 2501.17195 • Published • 35 -
WILDCHAT-50M: A Deep Dive Into the Role of Synthetic Data in Post-Training
Paper • 2501.18511 • Published • 20 -
Preference Leakage: A Contamination Problem in LLM-as-a-judge
Paper • 2502.01534 • Published • 40 -
NaturalReasoning: Reasoning in the Wild with 2.8M Challenging Questions
Paper • 2502.13124 • Published • 6 -
On Teacher Hacking in Language Model Distillation
Paper • 2502.02671 • Published • 18 -
MAGA: MAssive Genre-Audience Reformulation to Pretraining Corpus Expansion
Paper • 2502.04235 • Published • 22 -
DuoGuard: A Two-Player RL-Driven Framework for Multilingual LLM Guardrails
Paper • 2502.05163 • Published • 23 -
SynthDetoxM: Modern LLMs are Few-Shot Parallel Detoxification Data Annotators
Paper • 2502.06394 • Published • 89 -
Teaching Language Models to Critique via Reinforcement Learning
Paper • 2502.03492 • Published • 24 -
Distillation Scaling Laws
Paper • 2502.08606 • Published • 48 -
mmE5: Improving Multimodal Multilingual Embeddings via High-quality Synthetic Data
Paper • 2502.08468 • Published • 15 -
CLIPPER: Compression enables long-context synthetic data generation
Paper • 2502.14854 • Published • 11 -
Can Large Language Models Detect Errors in Long Chain-of-Thought Reasoning?
Paper • 2502.19361 • Published • 28 -
PlanGEN: A Multi-Agent Framework for Generating Planning and Reasoning Trajectories for Complex Problem Solving
Paper • 2502.16111 • Published • 9 -
VEM: Environment-Free Exploration for Training GUI Agent with Value Environment Model
Paper • 2502.18906 • Published • 12 -
Predictive Data Selection: The Data That Predicts Is the Data That Teaches
Paper • 2503.00808 • Published • 56 -
Process-based Self-Rewarding Language Models
Paper • 2503.03746 • Published • 39 -
KodCode: A Diverse, Challenging, and Verifiable Synthetic Dataset for Coding
Paper • 2503.02951 • Published • 33 -
LADDER: Self-Improving LLMs Through Recursive Problem Decomposition
Paper • 2503.00735 • Published • 23 -
Cognitive Behaviors that Enable Self-Improving Reasoners, or, Four Habits of Highly Effective STaRs
Paper • 2503.01307 • Published • 38 -
Sim-to-Real Reinforcement Learning for Vision-Based Dexterous Manipulation on Humanoids
Paper • 2502.20396 • Published • 15 -
TTRL: Test-Time Reinforcement Learning
Paper • 2504.16084 • Published • 120 -
Self-Taught Self-Correction for Small Language Models
Paper • 2503.08681 • Published • 15 -
MathFusion: Enhancing Mathematic Problem-solving of LLM through Instruction Fusion
Paper • 2503.16212 • Published • 25 -
MetaLadder: Ascending Mathematical Solution Quality via Analogical-Problem Reasoning Transfer
Paper • 2503.14891 • Published • 22 -
Temporal Consistency for LLM Reasoning Process Error Identification
Paper • 2503.14495 • Published • 11 -
OpenVLThinker: An Early Exploration to Complex Vision-Language Reasoning via Iterative Self-Improvement
Paper • 2503.17352 • Published • 24 -
Judge Anything: MLLM as a Judge Across Any Modality
Paper • 2503.17489 • Published • 23 -
JudgeLRM: Large Reasoning Models as a Judge
Paper • 2504.00050 • Published • 62 -
Inference-Time Scaling for Generalist Reward Modeling
Paper • 2504.02495 • Published • 56 -
Unicorn: Text-Only Data Synthesis for Vision Language Model Training
Paper • 2503.22655 • Published • 39 -
WikiVideo: Article Generation from Multiple Videos
Paper • 2504.00939 • Published • 37 -
RIG: Synergizing Reasoning and Imagination in End-to-End Generalist Policy
Paper • 2503.24388 • Published • 30 -
YourBench: Easy Custom Evaluation Sets for Everyone
Paper • 2504.01833 • Published • 22 -
OpenCodeReasoning: Advancing Data Distillation for Competitive Coding
Paper • 2504.01943 • Published • 15 -
GenPRM: Scaling Test-Time Compute of Process Reward Models via Generative Reasoning
Paper • 2504.00891 • Published • 14 -
DASH: Detection and Assessment of Systematic Hallucinations of VLMs
Paper • 2503.23573 • Published • 12 -
ActionStudio: A Lightweight Framework for Data and Training of Large Action Models
Paper • 2503.22673 • Published • 12 -
T1: Tool-integrated Self-verification for Test-time Compute Scaling in Small Language Models
Paper • 2504.04718 • Published • 42 -
MM-IFEngine: Towards Multimodal Instruction Following
Paper • 2504.07957 • Published • 35 -
MegaMath: Pushing the Limits of Open Math Corpora
Paper • 2504.02807 • Published • 34 -
Agentic Knowledgeable Self-awareness
Paper • 2504.03553 • Published • 27 -
SoTA with Less: MCTS-Guided Sample Selection for Data-Efficient Visual Reasoning Self-Improvement
Paper • 2504.07934 • Published • 20 -
APIGen-MT: Agentic Pipeline for Multi-Turn Data Generation via Simulated Agent-Human Interplay
Paper • 2504.03601 • Published • 17 -
SkillWeaver: Web Agents can Self-Improve by Discovering and Honing Skills
Paper • 2504.07079 • Published • 12 -
CLIMB: CLustering-based Iterative Data Mixture Bootstrapping for Language Model Pre-training
Paper • 2504.13161 • Published • 93 -
xVerify: Efficient Answer Verifier for Reasoning Model Evaluations
Paper • 2504.10481 • Published • 84 -
Iterative Self-Training for Code Generation via Reinforced Re-Ranking
Paper • 2504.09643 • Published • 34 -
Heimdall: test-time scaling on the generative verification
Paper • 2504.10337 • Published • 33 -
A Strategic Coordination Framework of Small LLMs Matches Large LLMs in Data Synthesis
Paper • 2504.12322 • Published • 28 -
Efficient Process Reward Model Training via Active Learning
Paper • 2504.10559 • Published • 13 -
MetaSynth: Meta-Prompting-Driven Agentic Scaffolds for Diverse Synthetic Data Generation
Paper • 2504.12563 • Published • 4 -
MIG: Automatic Data Selection for Instruction Tuning by Maximizing Information Gain in Semantic Space
Paper • 2504.13835 • Published • 38 -
Process Reward Models That Think
Paper • 2504.16828 • Published • 18