SuperScout is presented, which routes after scouting the repository: a 7B searcher, SuperScout-7B, first explores the repository and produces a structured handoff whose reproduction claims are sandbox-verified, with false claims stripped before delivery.
Abstract
Frontier language models can resolve repository-level software issues, but each attempt is expensive, and existing routers select a model from the issue text alone. We present SuperScout, which routes after scouting the repository: a 7B searcher, SuperScout-7B, first explores the repository and produces a structured handoff whose reproduction claims are sandbox-verified, with false claims stripped before delivery. The searcher's hidden states, together with the task text, then feed a resume-based router that dispatches the task to one of four frontier fixers. Adding a new fixer requires no retraining. On the full Python slice of SWE-bench Pro (266 tasks) under the benchmark's official capped budget tier, SuperScout matches the best single model's solve rate (159 of 266 for SuperScout, 158 for the best model) at about a fifth of the total cost per solve, and the reported configuration sits above the random traffic-splitting baseline. A no-router ablation, always the cheapest fixer with the handoff, ties the routed system on this benchmark, so the handoff rather than the routing decision carries the result. A paired calibration study points to the mechanism: the handoff appears to redistribute rather than add solving ability, lifting the three cheaper fixers while slightly hurting the strongest, though at $N=99$ the per-fixer effects are directional only; the searcher's hidden states improve cost routing on the calibration labels while the handoff's own text does not. The searcher's compute adds less than half a cent of GPU time per task.
A paired benchmark that tests whether LLM agents exhibit conscious allocation behavior under a fixed budget in two contexts: an abstract text-based formulation and a code-construction task finds that every frontier model testedacts near-optimally in the abstract framing but fails to transfer this ability to script-writing.
CRM+RCCR, an architecture-agnostic cost-aware objective that encodes cost preference into continuous relevance targets through per-pair independent scoring, eliminating multi-positive dilution while regularizing queries with similar routing preferences to be closer in the routing space.
Repository exploration is a distinct and costly stage of coding-agent pipelines: before generating a patch, an agent must identify which repository files are likely to matter. We study whether this stage can be delegated to lower-cost models while retaining useful localization quality. Using our IssueLoc-Bench, we evaluate five explorer models under the same read-only interactive interface on 499 SWE-bench Verified-derived tasks and 500 tasks from 153 additional repositories. We measure early candidate discovery, top-three gold-file coverage, strict file-set recovery, agent time, and token usage, with paired instance-level uncertainty and repository-clustered sensitivity analysis. The highest-quality explorer leads across localization metrics, but substantially cheaper operating points emerge: depending on the model and evaluation arm, lower-cost explorers retain approximately 78-94% of the reference Hit@3 and 73-92% of its F1 while reducing mean agent time by 41-88% and token usage by 84-95%. The preferred operating point depends on how localization is consumed downstream: ranking and coverage metrics characterize recoverable candidate handoffs, whereas F1 and exact match characterize restrictive file gates. These results support treating repository exploration as an independently measurable and budgetable stage of modular coding agents, with explorer selection guided by the downstream handoff contract.
This work compares configured Loreley QD, sequential champion editing, and independent root proposals in a matched Zstandard experiment and finds that Sequential had the highest observed 48-job mean and median and established a QD advantage.
A common measurement protocol and hybrid evaluation of four router implementations across RouterBench, BFCL v4, tau2-bench, and WebArena shows that, under these configurations and controls, observed gains track selected-tier composition more closely than demonstrated task-specific targeting.
In production, agentic systems answer questions over data that lives in several places and keeps changing: licenses are reassigned, users offboarded, prices changed, contracts renewed. Existing retrieval benchmarks freeze the data, so they cannot ask whether an agent's answer is still true, only whether it found the right passage. We present ChurnBench, an open-source benchmark that generates a four-source enterprise data fabric as a timeline rather than a snapshot. Every change is written to an append-only ground-truth ledger, and gold answers are computed from that ledger, never from the live stores. An answer that was correct when its data was retrieved but wrong when evaluated is therefore detected and labeled a freshness error, distinct from a reasoning error; we validate this by resolving ground truth at both timestamps for every case reported. Using the instrument, we find that when a system refreshes on a schedule, cache age does not predict staleness. Across cache ages of 1, 14, and 28 days, freshness errors were 7, 4, and 4, because scheduled refresh bounds staleness by time-to-live, and no TTL lapse was observed in any window. A controlled ablation confirms the mechanism: disabling tiered refresh raises freshness errors from 4 to 45 at 28 days and leaves them identical at one day. The variable a drift benchmark should sweep is therefore TTL configuration against each entity's rate of change, not drift-window length. ChurnBench, the evaluation harness, and all per-error data are released open source.
Unknown authors· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
