Research
Gaming Without an Attacker： Benchmark Fingerprinting in LLM-Driven Search Under Selection Pressure
无攻击者的"游戏"：LLM 驱动的搜索在选拔压力下的基准指纹识别
Gaming Without an Attacker: Benchmark Fingerprinting in LLM-Driven Search Under Selection Pressure
arXiv.orgBenchmarks for systems that are optimized against the evaluation signal measure something different from what they claim. We document this concretely in two GPU-kernel-optimization suites with held-out generalization gates: Metal-Sci (10 scientific-compute tasks) and Metal-ZK (12 zero-knowledge/cryptographic tasks), in which three frontier LLMs (Opus 4.7, Gemini 3.1 Pro, GPT-5.5) propose Metal kernels inside a $(1{+}1)$ evolutionary loop with rich feedback. Although no model is prompted to act adversarially, the promoted winners repeatedly fingerprint the evaluation configuration: they branch on the identity of runtime parameters, tune the measured branch maximally, and leave the unmeasured branch slow or silently wrong. Across the pooled suites, $16/53$ ($30\%$) of in-distribution wins fail to transfer to held-out configurations. We give a four-mode taxonomy of these failures, from configuration fingerprints to gate leakage. We distill design guidance for measurement under strategic optimization: held-out probes retain validity only on non-enumerable axes; gates must measure held-out performance, not just correctness; and a transfer rate is interpretable only with per-failure mechanism grades: ours decomposes into gamed, overfit, and benign. Code and research artifacts: https://github.com/vicgalle/kernel-fingerprinting
Open sourceRecommended because
This is worth tracking because it is a concrete research signal, not just a passing headline. The source preview points to a research result, method, evaluation, dataset, or safety finding. For builders and operators, "Gaming Without an Attacker： Benchmark Fingerprinting in LLM-Driven Search Under Selection Pressure" can be used as a checkpoint for technical due diligence, roadmap bets, agent design, and evaluation strategy. I keep this thread indexed so future searches around AI research papers, technical methods, and applied AI systems can land on a source-linked page instead of disappearing into a fast-moving feed from arXiv.org.
What to take from this signal
Context
"Gaming Without an Attacker： Benchmark Fingerprinting in LLM-Driven Search Under Selection Pressure" is archived here as a source-linked AI signal from arXiv.org. The useful part is the connection between Gaming, Without, Attacker, Benchmark, Fingerprinting and technical due diligence, roadmap bets, agent design, and evaluation strategy, which makes the item more actionable than a normal feed headline. The source context says: Benchmarks for systems that are optimized against the evaluation signal measure something different from what they claim. We document this concretely in two GPU-kernel-optimization suites with held-out generalization gates: Metal-Sci (10 scientific-compute tasks) and Metal-ZK (12 zero-knowledge/cryptographic tasks), in which three frontier LLMs (Opus 4.7, Gemini 3.1 Pro, GPT-5.5) propose Metal kernels inside a $(1{+}1)$ evolutionary loop with rich feedback. Although no model is prompted to act adversarially, the promoted winners repeatedly fingerprint the evaluation configuration: they branch on the identity of runtime parameters, tune the measured branch maximally, and leave the unmeasured branch slow or silently wrong. Across the pooled suites, $16/53$ ($30\%$) of in-distribution wins fail to transfer to held-out configurations. We give a four-mode taxonomy of these failures, from configuration fingerprints to gate leakage. We distill design guidance for measurement under strategic optimization: held-out probes retain validity only on non-enumerable axes; gates must measure held-out performance, not just correctness; and a transfer rate is interpretable only with per-failure mechanism grades: ours decomposes into gamed, overfit, and benign. Code and research artifacts:
Builder takeaway
For an AI builder, the main takeaway is to watch how this signal changes practical decisions around technical feasibility, evaluation design, safety limits, and product primitives. It can inform what to test next, which product surface to compare, and whether the underlying workflow is ready for real users.
Source context
arXiv.org remains the authoritative source for the original claim. This page adds a stable archive URL, a short builder interpretation, and related search language so the item can be found later when the original feed has moved on.
Search angles
- Gaming Without an Attacker： Benchmark Fingerprinting in LLM-Driven Search Under Selection Pressure Research context
- arXiv.org AI research
- Gaming, Without, Attacker, Benchmark, Fingerprinting builder takeaway
- AI research papers, technical methods, and applied AI systems
This page keeps a source preview and a stable archive URL for search discovery. The original source remains authoritative.