conftrace_

recency bia

8 papers

Explore in graph

Co-occurring keywords · papers mentioning both

large language model (2) language model (2) proof interpretation (1) reasoning reliability (1) nlp research (1) logit alignment (1) anchoring bia (1) decoding speed (1) sentential gate (1) proof reasoning (1)

Papers

Citation Amnesia: On The Recency Bias of NLP and Other Academic Fields COLING 2025 Linear Recency Bias During Training Improves Transformers’ Fit to Reading Times COLING 2025 Maintaining Fairness in Logit-based Knowledge Distillation for Class-Incremental Learning AAAI 2025 ProofTeller: Exposing recency bias in LLM reasoning and its side effects on communication IJCNLP 2025 Understanding and Mitigating Bottlenecks of State Space Models through the Lens of Recency and Over-smoothing ICLR 2025 MAGNIFICo: Evaluating the In-Context Learning Ability of Large Language Models to Generalize to Novel Interpretations EMNLP 2023 Modeling Context With Linear Attention for Scalable Document-Level Translation EMNLP 2022 Natural Analysts in Adaptive Data Analysis ICML 2019