DocInsights 2026 at EMNLP: keynotes by Dan Roth and Desmond Elliott, 48 accepted papers on document AI
keviv9 · x · 2026-10-24
The DocInsights 2026 workshop at EMNLP (Oct 29, Budapest) has published its full program: keynotes by Steffen Eger (UTN), Desmond Elliott (University of Copenhagen, on multilingual language modelling from pixels), and Dan Roth (UPenn/Oracle); shared tasks DocSem (evidence-grounded quantitative reasoning over documents) and Dr.DocBench (expert-level document parsing); and 48 accepted poster papers covering document parsing, multimodal document QA, low-cost page-stream segmentation, AdaptiveDocRAG, and accuracy-energy trade-offs for small local models. A useful agenda for anyone working on document understanding and information extraction.
More from Research
- Action learning: score whole trajectories with no forces or simulator, 2000× lower error — LeoTZ03 · 2026-10-24
- BeNice! method shares KV-cache across loops, scaling looped LLM depth without extra memory — stochasticchasm · 2026-10-24
- 215-state Turing machine proven to halt iff ZFC is inconsistent, verified in Metamath Zero — teortaxesTex · 2026-10-24
- OmniRay from JHU hits SOTA camera-agnostic monocular 3D reconstruction from one image — kwangmoo_yi · 2026-10-24
- AWS multi-turn RL guide: $339 lifts gpt-oss-20b Wordle solve rate 61%→84% — AWS ML Blog · 2026-10-24
- Eric Horvitz spotlights Stanford's tool-making research philosophy behind optogenetics — erichorvitz · 2026-10-24