Lune

Profiles2021-2026

Tao Gui

Fudan University

69Papers
9Last author
65Unique co-authors
0First author
0Solo

First and last author come from byline position, which some venues assign alphabetically.

Recent papers69

  1. BAPO: Stabilizing Off-Policy Reinforcement Learning for LLMs via Balanced Policy Optimization with Adaptive ClippingAuthor 19 of 21ICLR 2026 · 54 citations
  2. AgentPRM: Process Reward Models for LLM Agents via Step-Wise Promise and ProgressAuthor 13 of 15WWW 2026 · 19 citations
  3. OctoBench: Benchmarking Scaffold-Aware Instruction Following in Repository-Grounded Agentic CodingAuthor 15 of 15ACL 2026 · 10 citations
  4. Outcome Accuracy is Not Enough: Aligning the Reasoning Process of Reward ModelsAuthor 11 of 16ACL 2026 · 7 citations
  5. Unlocking the Essence of Beauty: Advanced Aesthetic Reasoning with Relative-Absolute Policy OptimizationAuthor 7 of 8ICLR 2026 · 6 citations
  6. Critique-RL: Training Language Models For Critiquing Through Two-Stage Reinforcement LearningAuthor 16 of 18ICLR 2026 · 4 citations
  7. Does Reinforcement Fine-Tuning Improve Generalization of LLM Agents? An Empirical StudyAuthor 12 of 14ICML 2026 · 3 citations
  8. Counteracting the Matthew Effect in Self-Improvement of LVLMs through Head-Tail Re-balancingAuthor 7 of 9ACL 2026 · 1 citation
  9. Feedback-Driven Tool-Use Improvements in Large Language Models via Automated Build EnvironmentsAuthor 9 of 11ACL Findings 2026 · 1 citation
  10. ABC-Bench: Benchmarking Agentic Backend Coding in Real-World DevelopmentAuthor 13 of 14ACL Findings 2026
  11. AgentGym-RL: An Open-Source Framework to Train LLM Agents for Long-Horizon Decision Making via Multi-Turn RLAuthor 20 of 24ICLR 2026
  12. AgentGym2: Benchmarking Large Language Model Agents in De-Idealized Real-World EnvironmentsAuthor 23 of 24ACL 2026
  13. AgentV-RL: Scaling Reward Modeling with Agentic VerifierAuthor 14 of 16ACL Findings 2026
  14. DARM: Distribution-Aware Reward Modeling by Alleviating Biases from Low Preference-Context Dependency DataAuthor 10 of 12ACL 2026
  15. MetaAct-RL: Training Language Models for Reasoning Through Meta-Action-Based Reinforcement LearningAuthor 18 of 20AAAI 2026
  16. MHA2MLA-VLM: Enabling DeepSeek's Economical Multi-Head Latent Attention Across Vision-Language ModelsAuthor 5 of 5AAAI 2026
  17. Muse: Towards Reproducible Long-Form Song Generation with Fine-Grained Style ControlAuthor 15 of 17ACL Findings 2026
  18. What Makes a Good Speech Tokenizer for LLM-Centric Speech Generation? A Systematic StudyAuthor 22 of 22AAAI 2026
  19. ToolHop: A Query-Driven Benchmark for Evaluating Large Language Models in Multi-Hop Tool UseAuthor 11 of 14ACL 2025 · 31 citations
  20. AgentGym: Evaluating and Training Large Language Model-based Agents across Diverse EnvironmentsAuthor 15 of 20ACL 2025 · 20 citations
  21. Measuring Data Diversity for Instruction Tuning: A Systematic Analysis and A Reliable MetricAuthor 9 of 11ACL 2025 · 15 citations
  22. Pre-Trained Policy Discriminators are General Reward ModelsAuthor 21 of 22NeurIPS 2025 · 13 citations
  23. EvaLearn: Quantifying the Learning Capability and Efficiency of LLMs via Sequential Problem SolvingAuthor 11 of 16NeurIPS 2025 · 11 citations
  24. Distill Visual Chart Reasoning Ability from LLMs to MLLMsAuthor 7 of 9EMNLP Findings 2025 · 6 citations
  25. Alleviating Shifted Distribution in Human Preference Alignment through Meta-LearningAuthor 11 of 13AAAI 2025 · 2 citations
  26. Analyzing the Effects of Supervised Fine-Tuning on Model Knowledge from Token and Parameter LevelsAuthor 6 of 10EMNLP 2025
  27. CritiQ: Mining Data Quality Criteria from Human PreferencesAuthor 11 of 11ACL 2025
  28. Governance in Motion: Co-evolution of Constitutions and AI models for Scalable SafetyAuthor 11 of 13EMNLP 2025
  29. Lost in the Context: Insufficient and Distracted Attention to Contexts in Preference ModelingAuthor 13 of 15ACL 2025
  30. Mitigating Object Hallucinations in MLLMs via Multi-Frequency PerturbationsAuthor 9 of 11EMNLP Findings 2025

Showing the 30 most recent indexed papers.

Dusk over the sea between two cliffs drawn in fine vertical lines