Weak-tier models often fail to invoke relevant harness artifacts during task-solving, with Qwen3-32B showing a 25% load rate against ~96% for strong models

Entities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.

• weak-tier models gain little, traced to two failure modes: failing to activate relevant harness artifacts and failing to follow them faithfully once activatedquote0.858
  Verbatim summary of weak-tier harness-benefit failure diagnosis from conclusion
• Even when the harness is loaded, weak-tier models fail to adhere to it due to weak instruction-following over long-horizon tasks, drifting more than four times more steeply than strong modelsclaim0.857
  Diagnosis of second failure mode explaining low harness-benefit for weak-tier models
• Strong-tier models benefit less from harness evolution because they already solve many tasks under the initial harness, leaving less room for improvement (ceiling effect)hypothesis0.823
  Explanation offered for why high-base-capability models show lower Δbenefit
• Weak-tier model deficits are not in task understanding but in protocol-level and procedural execution: they identify the right skill but cannot operate under itclaim0.807
  Diagnostic claim from case studies of activation and adherence failures in Qwen3-32B
• On SWE-bench, harness-benefit peaks at Qwen3-235B (19.3 pp), while weaker Qwen3-32B gains only 4.4 pp and stronger Opus 4.6 gains only 2.6 ppfinding0.761
  Core finding demonstrating non-monotonic relationship between base capability and harness-benefit
• Qwen3-32B adherence drops from 0.52 after harness loading to 0.13 at final validation (drift of -0.39)finding0.760
  Demonstrates long-horizon instruction-following bottleneck for weak-tier models
• does a model's base capability in task-solving predict its capabilities in harness self-evolution?question0.759
  Central framing question motivating the paper's capability decomposition
• QwQ and Qwen models have been extensively post-trained to excel at single-step tasks, causing degradation in long multi-turn interactions.hypothesis0.757
  Proposed explanation for why single-turn reformulation improves performance: models' training distribution is concentrated on single-turn reasoning.