AI & ML
The Missing Complement: State-Conditioned Minimal Sufficient Evidence for Coding Agents
arXiv cs.IR0 views
arXiv:2609.20050v3 Announce Type: replace Abstract: Retrieval assembles repository context by ranking passages for relevance to the current query. A coding agent halfway through an issue has already read much of what such a ranker returns. Relevance is scored per passage, but sufficiency belongs to the set: independently scored passages can fill the budget with support for one requirement while another goes unmet. We formulate state-conditioned minimal sufficient evidence recovery: given a captured agent state, recover a compact evidence combination supplying what its next decision still lacks. SERBench measures this on 500 held-out states from 45 repositories, recording what the agent has seen, crediting only sets that satisfy every annotated evidence requirement of the current decision, and separating set recovery from candidate discovery. MSS-Complement treats acquisition as set construction, not ranking. Three semantic calls propose a jointly sufficient set, search for what it lacks, and return 4-8 intact source units within 6,144 tokens. One configuration, fixed on calibration data, recovers a complete set for 73.0% of those states at five items and 80.6% at eight, against 61.4% and 72.4% for Qwen3 embedding with reranking. A matched control ranking by similarity alone recovers fewer complete sets, placing the margin over it in the set-level policy, not the computation. The lead persists from frozen repository source with no gold-derived pool. On AMA-Bench it answers from a 76.2% smaller answer prompt, with accuracy 2.08 points above that benchmark's own memory agent. Removing one required group from a complete set costs repair-localization precision under two executors. Retrieval for agents is better posed as recovering what a decision lacks than re-ranking what an issue resembles.
Source:https://arxiv.org/abs/2609.20050PreviousBeyond Dimensionality: Domain Shift and Catastrophic Forgetting in Single- vs. Multi-Vector RetrievalNextKeep it Simple! A Multiple-Choice Task can Benefit Generative Recommendation
Related posts
Hydra · Coeziune OM–IA — the complete DOI registry (2026-11-20): 8 published works
0dev.to OpenSource
Stop Renting Cloud GPUs & VPS: 4 Legitimate Free Compute Tiers in 2026 (Plus Billing Pitfalls)
0dev.to OpenSource
How to Decode a Wall of Wingdings Without a Key
0dev.to Programming
Beyond the Hype: How Modern Businesses Can Actually Benefit from AI Today
0dev.to Programming
How to Free Up Space on Android Without Deleting Anything
0dev.to Programming

All my git repos in one file: manage them day to day, rebuild them on a new machine
0dev.to Git