Profiles2021-2026
Yong Liu
Renmin University of China
48Papers
34Last author
59Unique co-authors
1First author
1Solo
First and last author come from byline position, which some venues assign alphabetically.
Recent papers48
- High Probability Bounds for Non-Convex Stochastic Optimization with MomentumAuthor 4 of 4ICLR 2026 · 100 citations
- Compositional Generalization from Learned Skills via CoT Training: A Theoretical and Structural Analysis for ReasoningAuthor 5 of 5ICLR 2026 · 6 citations
- Beyond Itinerary Planning - A Real-World Benchmark for Multi-Turn and Tool-Using Travel TasksAuthor 8 of 8ACL 2026 · 4 citations
- Weak-to-Strong Generalization via Bregman Bias–Variance DecompositionAuthor 4 of 4ICML 2026 · 4 citations
- No More Stale Feedback: Co-Evolving Critics for Open-World Agent LearningAuthor 10 of 10ACL 2026 · 3 citations
- ReasonAny: Incorporating Reasoning Capability to Any Model via Simple and Effective Model MergingAuthor 5 of 6ACL 2026 · 1 citation
- Demystifying Reasoning Dynamics with Mutual Information: Thinking Tokens are Information Peaks in LLM ReasoningAuthor 5 of 6NeurIPS 2025 · 63 citations
- Towards Reward Fairness in RLHF: From a Resource Allocation PerspectiveAuthor 5 of 5ACL 2025 · 10 citations
- Do not Abstain! Identify and Solve the UncertaintyAuthor 7 of 7ACL 2025 · 8 citations
- Exploring the Limitations of Mamba in COPY and CoT ReasoningAuthor 3 of 3EMNLP 2025 · 6 citations
- SSTAG: Structure-Aware Self-Supervised Learning Method for Text-Attributed GraphsAuthor 5 of 8NeurIPS 2025 · 5 citations
- Revisiting Chain-of-Thought Prompting: Zero-shot Can Be Stronger than Few-shotAuthor 8 of 8EMNLP Findings 2025 · 4 citations
- Stability and Generalization of Zeroth-Order Decentralized Stochastic Gradient Descent with Changing TopologyAuthor 7 of 7AAAI 2025 · 3 citations
- The Tug of War Within: Mitigating the Fairness-Privacy Conflicts in Large Language ModelsAuthor 4 of 5ACL 2025 · 3 citations
- Stability and Sharper Risk Bounds with Convergence Rate Õ(1/n2)Author 4 of 4NeurIPS 2025 · 2 citations
- Towards a Theoretical Understanding of Synthetic Data in LLM Post-Training: A Reverse-Bottleneck PerspectiveAuthor 2 of 2ICLR 2025 · 1 citation
- ADePT: Adaptive Decomposed Prompt Tuning for Parameter-Efficient Fine-tuningAuthor 3 of 3ICLR 2025
- Rethinking External Slow-Thinking: From Snowball Errors to Probability of Correct ReasoningAuthor 3 of 3ICML 2025
- Towards Auto-Regressive Next-Token Prediction: In-context Learning Emerges from GeneralizationAuthor 4 of 4ICLR 2025
- Understanding Model Ensemble in Transferable Adversarial AttackAuthor 4 of 4ICML 2025
- Neural Retrievers are Biased Towards LLM-Generated ContentAuthor 6 of 9KDD 2024 · 26 citations
- Towards Understanding How Transformers Learn In-context Through a Representation Learning LensAuthor 2 of 2NeurIPS 2024 · 26 citations
- ASWT-SGNN: Adaptive Spectral Wavelet Transform-Based Self-Supervised Graph Neural NetworkAuthor 3 of 4AAAI 2024 · 14 citations
- WaveNet: Tackling Non-stationary Graph Signals via Graph Spectral WaveletsAuthor 9 of 9AAAI 2024 · 9 citations
- FedNS: A Fast Sketching Newton-Type Algorithm for Federated LearningAuthor 2 of 4AAAI 2024 · 7 citations
- Enhancing In-Context Learning Performance with just SVD-Based Weight Pruning: A Theoretical PerspectiveAuthor 4 of 4NeurIPS 2024 · 4 citations
- High-Dimensional Analysis for Generalized Nonlinear Regression: From Asymptotics to AlgorithmAuthor 2 of 3AAAI 2024 · 4 citations
- Perfect Alignment May be Poisonous to Graph Contrastive LearningAuthor 3 of 3ICML 2024 · 4 citations
- Algorithmic Stability Unleashed: Generalization Bounds with Unbounded LossesAuthor 3 of 3ICML 2024 · 3 citations
- Concentration Inequalities for General Functions of Heavy-Tailed Random VariablesAuthor 2 of 2ICML 2024 · 1 citation
Showing the 30 most recent indexed papers.
