学者档案2021-2026
Yali Du
The Alan Turing Institute
25论文数
6末位作者
67合作者人数
0一作
0独著
一作与末位作者按论文署名顺序统计,部分顶会采用姓氏字母顺序署名。
近期论文25
- SocialJax: An Evaluation Suite for Multi-agent Reinforcement Learning in Sequential Social Dilemmas第 6 位,共 6 位作者ICLR 2026 · 11 次引用
- BRIDGE: Bi-level Reinforcement Learning for Dynamic Group Structure in Coalition Formation Games第 6 位,共 6 位作者ICLR 2026
- Entropy-Aware Reshaping of Reinforcement Signals for Multi-Answer Reasoning第 4 位,共 5 位作者ACL Findings 2026
- Is Pure Exploitation Sufficient in Exogenous MDPs with Linear Function Approximation?第 4 位,共 4 位作者ICLR 2026
- NOVER: Incentive Training for Language Models via Verifier-Free Reinforcement Learning第 5 位,共 6 位作者EMNLP 2025 · 4 次引用
- Abstract Counterfactuals for Language Model Agents第 3 位,共 5 位作者NeurIPS 2025 · 3 次引用
- Self-Verifying Reflection Helps Transformers with CoT Reasoning第 6 位,共 7 位作者NeurIPS 2025 · 3 次引用
- ATLAS: Agent Tuning via Learning Critical Steps第 4 位,共 6 位作者ACL Findings 2025 · 2 次引用
- Spiral of Silence in Large Language Model Agents第 6 位,共 8 位作者EMNLP Findings 2025 · 1 次引用
- CODI: Compressing Chain-of-Thought into Continuous Space via Self-Distillation第 5 位,共 6 位作者EMNLP 2025
- On the Optimization Landscape of Low Rank Adaptation Methods for Large Language Models第 2 位,共 4 位作者ICLR 2025
- Self-Guiding Exploration for Combinatorial Problems第 2 位,共 4 位作者NeurIPS 2024 · 27 次引用
- STAS: Spatial-Temporal Return Decomposition for Solving Sparse Rewards Problems in Multi-agent Reinforcement Learning第 4 位,共 4 位作者AAAI 2024 · 11 次引用
- PEARL: Zero-shot Cross-task Preference Alignment and Robust Reward Learning for Robotic Manipulation第 2 位,共 5 位作者ICML 2024 · 10 次引用
- Policy Learning from Tutorial Books via Understanding, Rehearsing and Introspecting第 3 位,共 7 位作者NeurIPS 2024 · 4 次引用
- TAPE: Leveraging Agent Topology for Cooperative Multi-Agent Policy Gradient第 5 位,共 5 位作者AAAI 2024 · 2 次引用
- Learning the Expected Core of Strictly Convex Stochastic Cooperative Games第 6 位,共 7 位作者NeurIPS 2024 · 1 次引用
- Invariant Learning via Probability of Sufficient and Necessary Causes第 4 位,共 8 位作者NeurIPS 2023 · 39 次引用
- An Efficient End-to-End Training Approach for Zero-Shot Human-AI Coordination第 6 位,共 6 位作者NeurIPS 2023 · 37 次引用
- Interpretable Reward Redistribution in Reinforcement Learning: A Causal Approach第 2 位,共 7 位作者NeurIPS 2023 · 32 次引用
- Reduced Policy Optimization for Continuous Control with Hard Constraints第 3 位,共 4 位作者NeurIPS 2023 · 10 次引用
- Rethinking Goal-Conditioned Supervised Learning and Its Connection to Offline RL第 6 位,共 9 位作者ICLR 2022 · 100 次引用
- Meta-Reward-Net: Implicitly Differentiable Reward Learning for Preference-based Reinforcement Learning第 3 位,共 4 位作者NeurIPS 2022 · 72 次引用
- Perceiving the World: Question-guided Reinforcement Learning for Text-based Games第 4 位,共 6 位作者ACL 2022 · 22 次引用
- Generalization in Text-based Games via Hierarchical Reinforcement Learning第 4 位,共 5 位作者EMNLP Findings 2021 · 13 次引用
