4B decision model trained on 117k examples matches Jev — except it can't detect sarcasm
SRSchmidgall · x · 2026-10-31
askalphaxiv, working with OpenResearch, trained a 4B decision model on over 117k decision-making examples. It performs practically identically to Jev across the board — except in sarcasm detection, a persistent failure mode. Notably, during evals the model generalized to unseen tasks like hate-speech severity, emotion detection, and intent classification.
More from Research
- AA-Omniscience v1.1 launches: Gemini 4 Carbon tops knowledge benchmark as some models drop 20+ points — ArtificialAnlys · 2026-10-31
- Santa Fe Institute publishes interactive history of agent-based modeling — round · 2026-10-31
- Peking University's DragMesh turns static 3D assets into articulated objects with one drag, accepted to IJCV — jiqizhixin · 2026-10-31
- Higher reasoning effort doesn't help AI agents catch silent failures, τ-bench test finds — TruthDeep8659 · 2026-10-31
- 'The Proof in the Code' named among 2026's best nonfiction: the story of Lean and human-AI collaboration — rbhar90 · 2026-10-31
- T-DIG uses an LLM to control any text-to-image model without weight access — simi_97k · 2026-10-31