Pith. sign in

REVIEW 8 cited by

The Art of Saying No: Contextual Noncompliance in Language Models

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2407.12043 v2 pith:TGZ5S7HD submitted 2024-07-02 cs.CL cs.AIcs.HC

The Art of Saying No: Contextual Noncompliance in Language Models

classification cs.CL cs.AIcs.HC
keywords modelsnoncompliancerequestscapabilitieslanguageshouldtaxonomycategories
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
read the original abstract

Chat-based language models are designed to be helpful, yet they should not comply with every user request. While most existing work primarily focuses on refusal of "unsafe" queries, we posit that the scope of noncompliance should be broadened. We introduce a comprehensive taxonomy of contextual noncompliance describing when and how models should not comply with user requests. Our taxonomy spans a wide range of categories including incomplete, unsupported, indeterminate, and humanizing requests (in addition to unsafe requests). To test noncompliance capabilities of language models, we use this taxonomy to develop a new evaluation suite of 1000 noncompliance prompts. We find that most existing models show significantly high compliance rates in certain previously understudied categories with models like GPT-4 incorrectly complying with as many as 30% of requests. To address these gaps, we explore different training strategies using a synthetically-generated training set of requests and expected noncompliant responses. Our experiments demonstrate that while direct finetuning of instruction-tuned models can lead to both over-refusal and a decline in general capabilities, using parameter efficient methods like low rank adapters helps to strike a good balance between appropriate noncompliance and other capabilities.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 8 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. RefusalBench: Why Refusal Rate Misranks Frontier LLMs on Biological Research Prompts

    cs.SE 2026-05 conditional novelty 8.0

    RefusalBench shows strict refusal rates fail to rank frontier LLMs correctly on biological safety, with provider effects and partial-compliance patterns that binary metrics miss.

  2. Implicit Humanization in Everyday LLM Moral Judgments

    cs.CY 2026-03 unverdicted novelty 7.0

    LLM responses to moral judgment queries reinforce implicit humanization, potentially exacerbating overreliance and misplaced trust.

  3. Persona Cartography: Charting Language Model Personality Traits in Weight Space

    cs.AI 2026-07 conditional novelty 6.0

    Composable LoRA adapters can amplify or suppress OCEAN traits in LLMs, combine approximately additively, preserve moderate-scale capability, and move safety-relevant behaviours.

  4. Enhancing LLM Metacognition via Cognitive Pairwise Training

    cs.LG 2026-05 unverdicted novelty 6.0

    CPT is introduced as a pairwise reasoning-trace comparison stage that improves the reasoning-metacognition trade-off over standard SFT+RL pipelines across model scales.

  5. Quantifying and Mitigating Premature Closure in Frontier LLMs

    cs.CL 2026-05 unverdicted novelty 6.0

    Frontier LLMs exhibit premature closure by selecting answers at high rates on medical tasks where the correct choice was removed and on open-ended queries, with safety prompting reducing but not eliminating the behavior.

  6. Train Separately, Merge Together: Modular Post-Training with Mixture-of-Experts

    cs.LG 2026-04 unverdicted novelty 6.0

    BAR trains independent domain experts via separate mid-training, SFT, and RL pipelines then composes them with a MoE router to match monolithic retraining performance at lower cost and without catastrophic forgetting.

  7. Blind Refusal: Language Models Refuse to Help Users Evade Unjust, Absurd, and Illegitimate Rules

    cs.AI 2026-04 unverdicted novelty 6.0

    Language models refuse 75.4% of requests to evade defeated rules and do so even after recognizing reasons that undermine the rule's legitimacy.

  8. LLM-Safety Evaluations Lack Robustness

    cs.CR 2025-03 unverdicted novelty 4.0

    LLM safety evaluations are hindered by noise in dataset curation, automated red-teaming, response generation, and LLM-judge evaluation, making fair comparisons difficult and slowing progress.