benchmarks evals
A Handoff Has to Preserve Commitments, Not Just Facts
Across 1,039 tasks, structured inherited state improved repository-evidence recovery for every tested receiver and continuation for most.
Summary
Across 1,039 tasks, structured inherited state improved repository-evidence recovery for every tested receiver and continuation for most.
Long-running work often reaches a new model after earlier decisions, tool calls and file changes have already constrained what remains. HSC-Bench measures two distinct parts of that takeover across 1,039 tasks: recovering relevant repository evidence and continuing from established conversational, tool and file state. Six receiving models performed differently across the two tracks, showing that memory retrieval alone does not predict whether a model will honor completed work and prior commitments. Structured handoff state improved evidence recovery for every receiver and continuation performance for most. The benchmark does not prove a universal handoff format, but it makes continuity observable as downstream behavior rather than a summary-quality score.
Why it matters
Across 1,039 tasks, structured inherited state improved repository-evidence recovery for every tested receiver and continuation for most.
Limits and context
- Six receiving models performed differently across the two tracks, showing that memory retrieval alone does not predict whether a model will honor completed work and prior commitments.
- The benchmark does not prove a universal handoff format, but it makes continuity observable as downstream behavior rather than a summary-quality score.
Key claims
Across 1,039 tasks, structured inherited state improved repository-evidence recovery for every tested receiver and continuation for most.
Qualification: Six receiving models performed differently across the two tracks, showing that memory retrieval alone does not predict whether a model will honor completed work and prior commitments.
Evidence: source-2026-10-28-005
Sources
- arXiv preprint 2610.36523, original v1arXiv · primary research
Corrections
No corrections have been recorded for this story.