collaborators

10 papers

cs.CL2026

SAS: Simple Attention Sparsification via End-to-End Optimization of Context Ranking

Zhiwei Li, Lei Zhu, Hao Gu +6

Post-training attention sparsification reduces the quadratic cumulative attention cost of pretrained Transformers by selecting a small set of context units (tokens or blocks) for e…

cs.LG2026

FlowBalance: Verifier-Grounded Self-Improvement from On-Policy Reasoning Experience

Zixun Huang, Kishan Panaganti, Haitao Mi +1

A reasoning model can improve from its own on-policy experience, but this inner loop is fragile: terminal verifiers provide reliable yet sparse supervision, while dense same-model…

cs.CL2026

Stabilizing Efficient Reasoning with Step-Level Advantage Selection

Han Wang, Xiaodong Yu, Jialian Wu +4

Large language models (LLMs) achieve strong reasoning performance by allocating substantial computation at inference time, often generating long and verbose reasoning traces. While…

cs.LG2026

Too Correct to Learn: Reinforcement Learning on Saturated Reasoning Data

Zhenwen Liang, Yujun Zhou, Sidi Lu +3

Reinforcement Learning (RL) enhances LLM reasoning, yet a paradox emerges as models scale: strong base models saturate standard benchmarks (e.g., MATH), yielding correct but homoge…

cs.AI2026

Training LLM Agents for Spontaneous, Reward-Free Self-Evolution via World Knowledge Exploration

Qifan Zhang, Dongyang Ma, Tianqing Fang +5

Most agents today ``self-evolve'' by following rewards and rules defined by humans. However, this process remains fundamentally dependent on external supervision; without human gui…

cs.AI2026

The Pensieve Paradigm: Stateful Language Models Mastering Their Own Context

Xiaoyuan Liu, Tian Liang, Dongyang Ma +4

In the world of Harry Potter, when Dumbledore's mind is overburdened, he extracts memories into a Pensieve to be revisited later. In the world of AI, while we possess the Pensieve-…