Copertina

Post-hoc Reward Calibration: A Case Study on Length Bias

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Huang, Zeyu, Qiu, Zihan, Wang, Zili, Ponti, Edoardo M., Titov, Ivan
Natura:	Preprint
Pubblicazione:	2024
Soggetti:	Artificial Intelligence Computation and Language
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

Blending Supervised and Reinforcement Fine-Tuning with Prefix Sampling
di: Huang, Zeyu, et al.
Pubblicazione: (2025)

Understanding Post-hoc Explainers: The Case of Anchors
di: Lopardo, Gianluigi, et al.
Pubblicazione: (2023)

Gender Bias in Explainability: Investigating Performance Disparity in Post-hoc Methods
di: Dhaini, Mahdi, et al.
Pubblicazione: (2025)

Emergent Communication Pretraining for Few-Shot Machine Translation
di: Li, Yaoyiran, et al.
Pubblicazione: (2020)

Probing the Emergence of Cross-lingual Alignment during LLM Training
di: Wang, Hetong, et al.
Pubblicazione: (2024)

Unlearning Traces the Influential Training Data of Language Models
di: Isonuma, Masaru, et al.
Pubblicazione: (2024)

M-Wanda: Improving One-Shot Pruning for Multilingual LLMs
di: Choenni, Rochelle, et al.
Pubblicazione: (2025)

Bias Is a Subspace, Not a Coordinate: A Geometric Rethinking of Post-hoc Debiasing in Vision-Language Models
di: Zhao, Dachuan, et al.
Pubblicazione: (2025)

Layerwise Recurrent Router for Mixture-of-Experts
di: Qiu, Zihan, et al.
Pubblicazione: (2024)

Scaling Sparse Fine-Tuning to Large Language Models
di: Ansell, Alan, et al.
Pubblicazione: (2024)

Controlling What You Share: Assessing Language Model Adherence to Privacy Preferences
di: Ramírez, Guillem, et al.
Pubblicazione: (2025)

What's New in My Data? Novelty Exploration via Contrastive Generation
di: Isonuma, Masaru, et al.
Pubblicazione: (2024)

Bootstrapping Action-Grounded Visual Dynamics in Unified Vision-Language Models
di: Qiu, Yifu, et al.
Pubblicazione: (2025)

Analysing Chain of Thought Dynamics: Active Guidance or Unfaithful Post-hoc Rationalisation?
di: Lewis-Lim, Samuel, et al.
Pubblicazione: (2025)

A Survey on Multilingual Large Language Models: Corpora, Alignment, and Bias
di: Xu, Yuemei, et al.
Pubblicazione: (2024)

AceMath: Advancing Frontier Math Reasoning with Post-Training and Reward Modeling
di: Liu, Zihan, et al.
Pubblicazione: (2024)

Spectral Editing of Activations for Large Language Model Alignment
di: Qiu, Yifu, et al.
Pubblicazione: (2024)

More Thinking, More Bias: Length-Driven Position Bias in Reasoning Models
di: Wang, Xiao
Pubblicazione: (2026)

Anthropomimetic Uncertainty: What Verbalized Uncertainty in Language Models is Missing
di: Ulmer, Dennis, et al.
Pubblicazione: (2025)

AdaThink-Med: Medical Adaptive Thinking with Uncertainty-Guided Length Calibration
di: Rui, Shaohao, et al.
Pubblicazione: (2025)

Post-hoc Utterance Refining Method by Entity Mining for Faithful Knowledge Grounded Conversations
di: Jang, Yoonna, et al.
Pubblicazione: (2024)

Act-Adaptive Margin: Dynamically Calibrating Reward Models for Subjective Ambiguity
di: Fang, Feiteng, et al.
Pubblicazione: (2025)

Mitigating Length Bias in RLHF through a Causal Lens
di: Kim, Hyeonji, et al.
Pubblicazione: (2025)

Detecting Prefix Bias in LLM-based Reward Models
di: Kumar, Ashwin, et al.
Pubblicazione: (2025)

Explaining Black-box Language Models with Knowledge Probing Systems: A Post-hoc Explanation Perspective
di: Zhao, Yunxiao, et al.
Pubblicazione: (2025)

Stacking Your Transformers: A Closer Look at Model Growth for Efficient LLM Pre-Training
di: Du, Wenyu, et al.
Pubblicazione: (2024)

One Bias After Another: Mechanistic Reward Shaping and Persistent Biases in Language Reward Models
di: Fein, Daniel, et al.
Pubblicazione: (2026)

Shared Doubt: Zero-shot Cross-Lingual Confidence Estimation for Language Models
di: Kyriakou, Athina, et al.
Pubblicazione: (2026)

Joint Localization and Activation Editing for Low-Resource Fine-Tuning
di: Lai, Wen, et al.
Pubblicazione: (2025)

Learn to Reason Efficiently with Adaptive Length-based Reward Shaping
di: Liu, Wei, et al.
Pubblicazione: (2025)

Scaling Unverifiable Rewards: A Case Study on Visual Insights
di: Gan, Shuyu, et al.
Pubblicazione: (2025)

Self-Improving World Modelling with Latent Actions
di: Qiu, Yifu, et al.
Pubblicazione: (2026)

The Effect of Model Size on LLM Post-hoc Explainability via LIME
di: Heyen, Henning, et al.
Pubblicazione: (2024)

CIRAG: Construction-Integration Retrieval and Adaptive Generation for Multi-hop Question Answering
di: Wei, Zili, et al.
Pubblicazione: (2026)

Understanding the Interplay of Scale, Data, and Bias in Language Models: A Case Study with BERT
di: Ali, Muhammad, et al.
Pubblicazione: (2024)

Mixtures of In-Context Learners
di: Hong, Giwon, et al.
Pubblicazione: (2024)

Guarded Repair for Harm-Aware Post-hoc Replacement of LLM Mathematical Reasoning
di: Xia, Haizhou
Pubblicazione: (2026)

AGR: Age Group fairness Reward for Bias Mitigation in LLMs
di: Cao, Shuirong, et al.
Pubblicazione: (2024)

Rewarding Doubt: A Reinforcement Learning Approach to Calibrated Confidence Expression of Large Language Models
di: Bani-Harouni, David, et al.
Pubblicazione: (2025)

DR.GAP: Mitigating Bias in Large Language Models using Gender-Aware Prompting with Demonstration and Reasoning
di: Qiu, Hongye, et al.
Pubblicazione: (2025)