agent-loop-skills-prompts-held-fixed-only-model-swapped CLM-0334observationsingle paperReviewed by AI
With the agent loop, skills and prompts held fixed and only the model swapped, models split into two clusters on operating a multi-step asynchronous ML workflow: Claude Sonnet 3.5-4.0, Gemini 2.5 and GPT-5 completed tasks at 93-100%, while GPT-4 scored 13% and others could not run the loop at all, hallucinating workflow ids or failing to wait for asynchronous jobs; on open-ended exploration the effect of a stressful, competitive prompt reversed by family, with Sonnet 4.0 producing the largest improvement of any configuration under stress and GPT-5 turning conservative and giving back most of its basic-prompt gains.
Capability: Using the tools it is given
Observed on
A-MLE harness in an industrial ads-ranking stack, 2026; task-completeness scores on tier-2 workflow execution and rMSE improvement on tier-3 exploration; one prompt pair (basic ver.
Sources
- A rare controlled model swap inside a real harness; small model set and a single prompt manipulation, so the stress-reversal finding is suggestive rather than established. The completeness split matches what harness builders report elsewhere.
Either way, bring evidence: a link, or what you saw.
Disagreeing is the most useful thing you can do here. Nothing on this page waits for a reviewer: the counts above move when evidence lands, and every source stays attached with who brought it.
Related claims
- When a model must anticipate, without execution feedback, what a change to its own runtime harness components will do, naming which components a lifecycle change can affect stays near ceiling as the number of interacting effects grows from 2 to 32, while predicting the resulting state, reasoning across teardown orders, and choosing minimal reconfigurations degrade sharply (e.g. GPT-5.6 Luna's reachable-condition Jaccard 91.7% to 14.1%, executed reconfiguration success 62.5% to 25.0%).Improving its own scaffolding from failures · reviewed by AI
- For a frozen-backbone multi-agent scientific discovery pipeline on HypoBench tasks, distilling reusable procedural skills separately per stage — from discriminator critique of a whole hypothesis batch for generation, and from execution outcomes compared to ground-truth results for testing — beats no skill, an LLM-written skill, and a hand-written skill, and removing each stage's feedback signal costs most of the gain.Improving its own scaffolding from failures · reviewed by AI
- For open-source LLM agents driving a ROS robot, having the model emit a pseudo-code tool-call plan up front, storing it in a persistent scratchpad, and then restricting each tool invocation to the plan's next step raises task success most on the longest multi-step tasks and for the smaller open models (GPT-OSS-20B on the hardest task rose from 35% with a persistent plan alone to 55% with the constraint), while a strong proprietary model gains little.Completing tasks that take many steps or many hours · reviewed by AI
- In multi-agent LLM workflows with a mixed-size model zoo, choosing which model instantiates each agent step-by-step from an online estimate of task-progress gain plus remaining time and cost budget keeps task quality near the best single-model policy at lower energy than one-shot query-level routing or stall-triggered escalation.Getting more from several models than from one · reviewed by AI
- On simple HLS kernel generation tasks (85 designs from CHStone, MachSuite, Polybench, Rosetta, C2HLSC), giving a modest open-source model (gpt-oss-120b) a shell with a C++ compiler for self-verification and iterative refinement saturates the benchmark at pass@10 (100% through synthesis) and exceeds 90% at pass@1, so this class of task no longer separates models; repeated sampling helps the weaker model far more (+60 points synthesis pass rate at k=10 for gpt-oss-20b vs +11 for the 120b).Generating and editing working code · reviewed by AI
Notes
Drafted from the paper by a model and filed unreviewed. Visible here so it can be read, not because anyone has vouched for it: it does not move any technique's standing and does not count toward the internal scorecard. Drafted confidence: low. Falsifier as drafted: Replication of the prompt-stress reversal on a public agentic benchmark with the same model pairs, or the cluster split disappearing once the weaker models are given the same retry and wait tooling.