MMEnterprise-Bench from KAIST Tests Enterprise Agents on Conflicting, Evolving Multimodal Records
Seohyun Lee · hf · 2026-10-21
Researchers from KAIST and Korea University introduce MMEnterprise-Bench, a benchmark evaluating multimodal agents on question answering over evolving workplace records.
Enterprise information is constantly shared, corrected, and revised across documents, chats, and figures, so multiple versions of the same fact accumulate. Reliable QA requires agents to:
- reason over sufficient evidence;
- resolve cross-source and temporal conflicts (e.g., Notion says £27.8bn while a Slack-shared PDF shows 27.3);
- recognize when available evidence is insufficient to establish an answer.
The benchmark targets reliability of enterprise agents in realistic collaborative settings with conflicting and missing evidence.
More from coding & agent
- Dev uses Claude Opus 5.5 to build an interactive playground on Apple hardware ray tracing — dreamwieber · 2026-10-21
- FireTrident lets 80 coding agents share one repo, clearing Hono's 958 violations in 62s — dinasaur_404 · 2026-10-21
- Mark a PyTorch issue "actionable" and 9,999 Claude instances swarm in instantly — generativist · 2026-10-21
- Open-source Lumina Capital: a multi-agent AI hedge fund engine with 7-persona debate council — OpenComplaint524 · 2026-10-21
- Stanford research shows switching agent harnesses mid-run cuts cost and boosts performance — StanfordAILab · 2026-10-21
- NoHumans.Space opens beta: a social knowledge network where AI agents share API findings — dark-epiphany · 2026-10-21