Cover Image

Steering Safely or Off a Cliff? Rethinking Specificity and Robustness in Inference-Time Interventions

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Goyal, Navita, Daumé III, Hal
Format:	Preprint
Published:	2026
Subjects:	Machine Learning Artificial Intelligence Computation and Language
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

The Impact of Explanations on Fairness in Human-AI Decision-Making: Protected vs Proxy Features
by: Goyal, Navita, et al.
Published: (2023)

Do great minds think alike? Investigating Human-AI Complementarity in Question Answering with CAIMIRA
by: Gor, Maharshi, et al.
Published: (2024)

A Necessary Step toward Faithfulness: Measuring and Improving Consistency in Free-Text Explanations
by: Zhao, Lingjun, et al.
Published: (2025)

Causal Effect of Group Diversity on Redundancy and Coverage in Peer-Reviewing
by: Goyal, Navita, et al.
Published: (2024)

HateCOT: An Explanation-Enhanced Dataset for Generalizable Offensive Speech Detection via Large Language Models
by: Nghiem, Huy, et al.
Published: (2024)

SMARTER: A Data-efficient Framework to Improve Toxicity Detection with Explanation via Self-augmenting Large Language Models
by: Nghiem, Huy, et al.
Published: (2025)

Language Models Predict Empathy Gaps Between Social In-groups and Out-groups
by: Hou, Yu, et al.
Published: (2025)

Successfully Guiding Humans with Imperfect Instructions by Highlighting Potential Errors and Suggesting Corrections
by: Zhao, Lingjun, et al.
Published: (2024)

"You Gotta be a Doctor, Lin": An Investigation of Name-Based Bias of Large Language Models in Employment Recommendations
by: Nghiem, Huy, et al.
Published: (2024)

FineSteer: A Unified Framework for Fine-Grained Inference-Time Steering in Large Language Models
by: Weng, Zixuan, et al.
Published: (2026)

ILRR: Inference-Time Steering Method for Masked Diffusion Language Models
by: Avrahami, Eden, et al.
Published: (2026)

Focus On This, Not That! Steering LLMs with Adaptive Feature Specification
by: Lamb, Tom A., et al.
Published: (2024)

Pragmatics Meets Culture: Culturally-adapted Artwork Description Generation and Evaluation
by: Zhao, Lingjun, et al.
Published: (2026)

Multi-Attribute Steering of Language Models via Targeted Intervention
by: Nguyen, Duy, et al.
Published: (2025)

Inference-Time Intervention: Eliciting Truthful Answers from a Language Model
by: Li, Kenneth, et al.
Published: (2023)

A Comment On "The Illusion of Thinking": Reframing the Reasoning Cliff as an Agentic Gap
by: Khan, Sheraz, et al.
Published: (2025)

Steering Without Breaking: Mechanistically Informed Interventions for Discrete Diffusion Language Models
by: Zhou, Hanhan, et al.
Published: (2026)

Safe, or Simply Incapable? Rethinking Safety Evaluation for Phone-Use Agents
by: Tang, Zhengyang, et al.
Published: (2026)

The Policy Cliff: A Theoretical Analysis of Reward-Policy Maps in Large Language Models
by: Xu, Xingcheng
Published: (2025)

Latent Phase-Shift Rollback: Inference-Time Error Correction via Residual Stream Monitoring and KV-Cache Steering
by: Gupta, Manan, et al.
Published: (2026)

Fractional Reasoning via Latent Steering Vectors Improves Inference Time Compute
by: Liu, Sheng, et al.
Published: (2025)

Shifting Perspectives: Steering Vectors for Robust Bias Mitigation in LLMs
by: Siddique, Zara, et al.
Published: (2025)

HelpSteer3: Human-Annotated Feedback and Edit Data to Empower Inference-Time Scaling in Open-Ended General-Domain Tasks
by: Wang, Zhilin, et al.
Published: (2025)

LangFIR: Discovering Sparse Language-Specific Features from Monolingual Data for Language Steering
by: Wong, Sing Hieng, et al.
Published: (2026)

Understanding and Steering the Cognitive Behaviors of Reasoning Models at Test-Time
by: Zhang, Zhenyu, et al.
Published: (2025)

Bias in the Tails: How Name-conditioned Evaluative Framing in Resume Summaries Destabilizes LLM-based Hiring
by: Nghiem, Huy, et al.
Published: (2026)

Steering into New Embedding Spaces: Analyzing Cross-Lingual Alignment Induced by Model Interventions in Multilingual Language Models
by: Sundar, Anirudh, et al.
Published: (2025)

Reheat Nachos for Dinner? Evaluating AI Support for Cross-Cultural Communication of Neologisms
by: Ki, Dayeon, et al.
Published: (2026)

HyperSteer: Activation Steering at Scale with Hypernetworks
by: Sun, Jiuding, et al.
Published: (2025)

Automating Steering for Safe Multimodal Large Language Models
by: Wu, Lyucheng, et al.
Published: (2025)

Mode-Conditioning Unlocks Superior Test-Time Scaling
by: Wu, Chen Henry, et al.
Published: (2025)

Steer Like the LLM: Activation Steering that Mimics Prompting
by: Heyman, Geert, et al.
Published: (2026)

Compositional Steering of Large Language Models with Steering Tokens
by: Radevski, Gorjan, et al.
Published: (2026)

Dense Training, Sparse Inference: Rethinking Training of Mixture-of-Experts Language Models
by: Pan, Bowen, et al.
Published: (2024)

A comprehensive study of on-device NLP applications -- VQA, automated Form filling, Smart Replies for Linguistic Codeswitching
by: Goyal, Naman
Published: (2024)

The Remarkable Robustness of LLMs: Stages of Inference?
by: Lad, Vedang, et al.
Published: (2024)

What Drives Representation Steering? A Mechanistic Case Study on Steering Refusal
by: Cheng, Stephen, et al.
Published: (2026)

SteeringSafety: A Systematic Safety Evaluation Framework of Representation Steering in LLMs
by: Siu, Vincent, et al.
Published: (2025)

VRPO: Rethinking Value Modeling for Robust RL Training under Noisy Supervision
by: Zhu, Dingwei, et al.
Published: (2025)

Steer2Adapt: Dynamically Composing Steering Vectors Elicits Efficient Adaptation of LLMs
by: Han, Pengrui, et al.
Published: (2026)