Pith. sign in

REVIEW 9 cited by

R-Tuning: Instructing Large Language Models to Say `I Don't Know'

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 2311.09677 v3 pith:EAXUTUXV submitted 2023-11-16 cs.CL

classification cs.CL
keywords knowledger-tuningabilityinstructionmodelmodelsquestionstuning
verification ladder T0 review T1 audit T2 compute T3 formal

Signed reviews

No signed human review yet.

0 comments
read the original abstract

Large language models (LLMs) have revolutionized numerous domains with their impressive performance but still face their challenges. A predominant issue is the propensity for these models to generate non-existent facts, a concern termed hallucination. Our research is motivated by the observation that previous instruction tuning methods force the model to complete a sentence no matter whether the model knows the knowledge or not. When the question is out of the parametric knowledge, it will try to make up something and fail to indicate when it lacks knowledge. In this paper, we present a new approach called Refusal-Aware Instruction Tuning (R-Tuning). This approach is formalized by first identifying the disparity in knowledge encompassed by pre-trained parameters compared to that of instruction tuning data. Then, we construct the refusal-aware data based on the knowledge intersection, to tune LLMs to refrain from responding to questions beyond its parametric knowledge. Experimental results demonstrate R-Tuning effectively improves a model's ability to answer known questions and refrain from answering unknown questions. Furthermore, when tested on out-of-domain datasets, the refusal ability was found to be a meta-skill that could be generalized to other tasks. Further analysis surprisingly finds that learning the uncertainty results in better calibration and an improved ability to estimate the uncertainty than uncertainty-based testing. Our code is available at https://github.com/shizhediao/R-Tuning.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 9 Pith papers

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

  1. Do LLM Recommenders Know When They're Hallucinating? Auditing Confidence Calibration in Catalog Faithfulness

    cs.IR 2026-08 conditional novelty 7.0 of 10

    Verbalized confidence from four zero-shot LLM recommenders is systematically under-confident and cannot separate correct items from catalog hallucinations, so confidence-gated abstention barely reduces hallucination.

  2. PhantomFill: When the Form Demands an Answer, Language Models Invent One

    cs.LG 2026-06 conditional novelty 7.0 of 10

    Required JSON fields make LLMs invent answers to unanswerable questions 100% of the time in ten of thirteen models, even when an 'insufficient evidence' escape exists.

  3. Prompt Compression via Activation Aggregation

    cs.CL 2026-07 conditional novelty 6.0 of 10

    A learned weighted sum of intermediate-layer activations compresses an instruction prompt into a single patch vector that, injected at an early layer, recovers task accuracy within ~2% of the full prompt.

  4. Writing Like the Best: Exemplar-Based Expository Text Generation

    cs.CL 2025-05 conditional novelty 6.0 of 10

    Exemplar-based expository generation: a recurrent plan-then-adapt LLM framework converts a source topic's exemplar into target-topic text by question transfer, retrieval, and confidence-gated answering.

  5. Structural Entropy Guided Agent for Detecting and Repairing Knowledge Deficiencies in LLMs

    cs.CL 2025-05 conditional novelty 6.0 of 10

    SENATOR guides a language model through a knowledge graph, measures its uncertainty with structural entropy, and fine-tunes it on synthetic data chosen to fix its weak spots, gaining up to 12 percent average relative ...

  6. Cannot or Should Not? Automatic Analysis of Refusal Composition in IFT/RLHF Datasets and Refusal Behavior of Black-Box LLMs

    cs.AI 2024-12 conditional novelty 6.0 of 10

    A unified 16-category refusal taxonomy with human and synthetic datasets and a low-cost classifier for auditing refusal behavior in LLMs.

  7. UAlign: Leveraging Uncertainty Estimations for Factuality Alignment on Large Language Models

    cs.CL 2024-12 conditional novelty 6.0 of 10

    UAlign improves LLM factuality alignment by adding predicted confidence and semantic entropy as input features to prompts and the reward model, helping the model answer known questions and refuse unknown ones.

  8. Inteligencia Artificial jur\'idica y el desaf\'io de la veracidad: an\'alisis de alucinaciones, optimizaci\'on de RAG y principios para una integraci\'on responsable

    cs.AI 2025-09 conditional novelty 4.0 of 10

    Legal AI hallucination persists in commercial RAG tools (17-34%+ of queries), so the report argues the fix is consultative, source-citing system design plus mandatory human oversight, not better generative models.

  9. Ask Good Questions for Large Language Models

    cs.CL 2025-08 unverdicted novelty 4.0 of 10

    The AGQ framework combines a concept-enhanced item response theory model with LLMs to generate guiding questions that adapt to a user's estimated knowledge gaps.

Pith tools