anchoring-bias

Tag

Cards List
#anchoring-bias

Prompt Framing Distorts Count-Based Evaluation of LLM Error Detection: Evidence from Numeric Anchoring

arXiv cs.CL · 2026-07-03 Cached

This paper introduces ErrorBench, a stress-test protocol demonstrating that numeric anchoring in prompts inflates count-based F1 scores for LLM error detection without improving span localization, making count-only evaluation unreliable.

0 favorites 0 likes
#anchoring-bias

Mitigating Anchoring Bias in LLM-Based Agents for Energy-Efficient 6G Autonomous Networks

arXiv cs.AI · 2026-06-18 Cached

This paper proposes a randomized anchoring strategy to mitigate anchoring bias in LLM-based agents for energy-efficient 6G autonomous networks, achieving up to 25% energy savings using a lightweight 1B-parameter model.

0 favorites 0 likes
#anchoring-bias

Localizing Anchoring Pathways in Language Models

arXiv cs.CL · 2026-06-12 Cached

This paper investigates how irrelevant numbers in prompts cause anchoring effects in language models and localizes the internal pathways carrying this signal using attribution-based circuit methods on Qwen and Llama models.

0 favorites 0 likes
← Back to home

Submit Feedback