AI
Omni-Decision: Evidence-Ledger Planning for Omni-Modal Agents
Publication: Hugging Face Daily Papers
Author: Not available in collected metadata
Source published: 23/09/2026, 20:00:00 UTC
Briefing created: 30/09/2026, 06:56:05 UTC
Based on a feed excerpt; full article content has not been acquired. Source evidence retrieved 30/09/2026, 06:18:24 UTC.
Briefings
Omni-Decision is an omni-modal agent that replaces noisy dialogue history with an explicit evidence ledger to improve planning. By filtering observations and recording missing/confirmed evidence, it achieves 81.4% accuracy on OmniGAIA at 43% of Gemini-3.1-Pro's cost per question, matching WorldSense long-video understanding performance.
Key points
- Omni-Decision replaces noisy multimodal observations with an explicit evidence ledger that records missing evidence, confirmed evidence, and conflicting records to support planning.
- Replacing the planner causes significantly larger performance loss than replacing the perception backend, indicating the planner is the primary bottleneck.
Context and limitations
Evidence is limited to the provided text; specific implementation details, exact training data distributions, and long-term stability results are not covered.
Generated by Bittrees-hosted qwen35-2b sha256:aaf42c8b7c3c on 30/09/2026, 06:56:05 UTC. Based on publisher-provided feed evidence; may not cover the full original article. Check the source for details.
View archived briefing on IPFS
bafybeifruqlqjfy2wwp4fvxvm5oyqx4pf2jflp33k2tj66yyejwlb7v6za