Harness invocation should be treated as a first-class learned skill and baked into agent training, as weak-tier models fail to load skills 75% of the time

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

• Harness-benefit is non-monotonic in base capability: weak-tier models benefit little, mid-tier models benefit most, and strong-tier models benefit less than mid-tierclaim0.789
  Second major claim of the paper, supported by Δbenefit measurements across six models on three benchmarks
• 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 modelsclaim0.760
  Diagnosis of second failure mode explaining low harness-benefit for weak-tier models
• weak-tier models gain little, traced to two failure modes: failing to activate relevant harness artifacts and failing to follow them faithfully once activatedquote0.757
  Verbatim summary of weak-tier harness-benefit failure diagnosis from conclusion
• Transformers almost surely maintain input-injectivity throughout training, not just at initialisationhypothesis0.746
  Conjecture supported by Nikolaou et al. 2025 for last-token hidden states
• what explains why weak-tier models with the most performance headroom benefit least from harness evolution?question0.744
  In-depth diagnostic question addressed by the two failure mode analysis
• 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)hypothesis0.744
  Explanation offered for why high-base-capability models show lower Δbenefit
• Transformers develop self-models through in-context learning, not just training data; even old base models without LLM-related text can bootstrap self-referential reasoning at runtime.claim0.742
  Antra's foundational claim about how introspection arises computationally rather than from memorised text.
• Current application of 15 properties to agent harness is metaphorical; operationalization into measurement is an open empirical question.claim0.739