Two heuristic code agents outperform most tested LLMs, and behavioural traces surface recurring LLM failure modes including overbidding, self-bidding, bankrupt TC initiation, and weak opponent-state adaptation.

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

• Behavioural traces surface recurring LLM failure modes including overbidding, self-bidding, bankrupt TC initiation, and weak opponent-state adaptation that never appear in code agents.claim0.917
  LLMs exhibit systematic errors that deterministic logic avoids.
• Two heuristic code agents outperform most tested LLMsclaim0.863
  author assertion that deterministic heuristics surpass many LLMs
• Two heuristic code agents (TrackerAgent and SetRaceAgent) outperform most tested LLMs.claim0.851
  Calibration that conditional logic can beat cost-efficient LLMs in this setting.
• Code agents operate on structured data with exact arithmetic, while LLMs must parse natural-language observations and track state across turns; some failures may partly reflect numerical parsing or working-memory limitationsclaim0.834
  discussion of potential confounds
• Conditional logic already suffices where LLMs still fail, as code agents avoid systematic failuresclaim0.821
  contrast between rule-based and LLM reasoning
• Overbid frequency, self-bidding rate, bankrupt-initiation patterns, and context-dependent offer calibration are failure modes invisible to both static evaluations and aggregate rankings like Eloclaim0.814
  key claim about the benchmark's unique diagnostic value
• Card-counting heuristics suffice to outperform most LLMs tested.claim0.809
  TrackerAgent's second-place ranking calibrates the benchmark and highlights LLM shortcomings.
• Do LLM failures in CATTLE TRADE reflect genuinely hard strategic problems or errors that novice humans also avoid?question0.787
  Open question about benchmarking against human players to calibrate difficulty.