10 papers
DocAtlas: Long-Document Understanding as Mutable-State Interaction
Hongchen Wei, Yuanzhe Wang, Bei Liu +8
Long-document understanding requires models to find and combine evidence across many pages, layouts, tables, figures, and charts. Existing retrieval-augmented systems usually selec…
XL-DocBench: Benchmarking Evidence-Grounded Extra-Long Document Understanding
Hongchen Wei, Yuanzhe Wang, Bei Liu +9
Real-world document tasks often ask professionals to answer questions from annual reports, regulations, clinical guidelines, and technical manuals that span hundreds or thousands o…
RESOURCE2SKILL: Distilling Executable Agent Skills from Human-Created Multimodal Resources
Yijia Fan, Zonglin Di, Zimo Wen +8
The paper introduces RESOURCE2SKILL, a framework that converts multimodal human-created resources such as tutorial videos, code repositories, and articles into executable skills or…
IndexMem: Learned KV-Cache Eviction with Latent Memory for Long-Context LLM Inference
Xintong Yang, Hao Gu, Binxing Xu +6
Large Language Models (LLMs) are increasingly expected to operate over long contexts, yet standard softmax attention incurs a KV cache that grows linearly with sequence length, qui…
Token Predictors Are Not Planners: Building Physically Grounded Causal Reasoners
Zheng Lu, Mingqi Gao, Qinlei Xie +8
Current benchmarks for embodied vision-language planning often favor linguistic next-token prediction over physically grounded next-state reasoning. This rewards models that mimic…
SkillOpt: Executive Strategy for Self-Evolving Agent Skills
Yifan Yang, Ziyang Gong, Weiquan Huang +12
Agent skills today are hand-crafted, generated one-shot, or evolved through loosely controlled self-revision, none of which behaves like a deep-learning optimizer for the skill, an…