Benchmark RadarAI BENCHMARK PROFILE
AnchorBench
AnchorBench evaluates anchoring effects in LLMs across multiple pathways and anchor relevance, measuring answer shifts through controlled prompts.
- Released
- 2026-08-14
- Readiness
- Paper only
- Primary field
- General AI
Why it matters
It addresses gaps in prior anchoring benchmarks by distinguishing irrelevant and plausible anchors and evaluating pathway dependence.
Motivation
The anchoring effect is a cognitive bias in which an initial reference value shifts a later judgment toward itself.
Primary resources
Benchmark Radar records only publicly supported details and links back to primary sources for verification.