Yi-Fan Yang

RCEdit-500K: Reference Completion for Image-Conditioned Image Editing

Reasoning-Informed Visual Editing

Large Multi-modality Models (LMMs) have made significant progress in visual understanding and generation, but still face challenges in visual editing, particularly in following complex instructions, preserving appearance consistency, and supporting flexible input formats. To study this gap, we introduce RISEBench, the...

Inspire: Benchmarking Scientific Literature Search for Open Research Problems

INSPIRE, a benchmark for evaluating agents that search prior literature to make progress on solution-redacted research problems, enables both end-to-end comparison and stage-resolved diagnosis in a setting where the agent must construct its own working criterion of relevance.

DENSE: Distilling Agent Trajectories into Evidence-Grounded Shortcut Trees for Self-Refinement

Online agent deployments accumulate execution trajectories at massive scale and behavioral diversity, for which predefined annotation criteria hardly exist. Extracting useful evidence therefore demands costly manual annotation or verifier signals that fail to scale, leaving valuable evidence buried among redundant, inc...

ReDeck: Step-Level Render-Grounded Refinement for Document-to-Slide Generation

ReDeck is proposed, a step-level render-grounded refinement framework that decomposes slide revision into atomic edit actions and returns renderer-derived observations after each step, turning refinement into"one edit, one observation."

XL-DocBench: Benchmarking Evidence-Grounded Extra-Long Document Understanding

XL-DocBench is introduced, a fully human-verified benchmark for extra-long document understanding, with 1,519 retained questions from six professional domains and contexts up to 2,303 pages, and fills a gap left by prior single-page, short multi-page, or text-only long-context benchmarks.

DocAtlas: Long-Document Understanding as Mutable-State Interaction

DocAtlas is presented, a system that treats long-document understanding as a mutable-state information-seeking process and instantiate DocAtlas as a mutable document harness: an external environment that determines what document information is searched, read, stored, reviewed, and shown to the model at each step.