Pith. sign in

REVIEW 2 cited by

Program-Aided Reasoners (better) Know What They 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 2311.09553 v1 pith:2TKJOA7B submitted 2023-11-16 cs.AI

classification cs.AI
keywords knowmodelscalibrationprogram-aidedreasonerstheywhataccuracy
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Prior work shows that program-aided reasoning, in which large language models (LLMs) are combined with programs written in programming languages such as Python, can significantly improve accuracy on various reasoning tasks. However, while accuracy is essential, it is also important for such reasoners to "know what they know", which can be quantified through the calibration of the model. In this paper, we compare the calibration of Program Aided Language Models (PAL) and text-based Chain-of-thought (COT) prompting techniques over 5 datasets and 2 model types: LLaMA models and OpenAI models. Our results indicate that PAL leads to improved calibration in 75% of the instances. Our analysis uncovers that prompting styles that produce lesser diversity in generations also have more calibrated results, and thus we also experiment with inducing lower generation diversity using temperature scaling and find that for certain temperatures, PAL is not only more accurate but is also more calibrated than COT. Overall, we demonstrate that, in the majority of cases, program-aided reasoners better know what they know than text-based counterparts.

Discussion (0). Sign in to comment.

Forward citations

Cited by 2 Pith papers

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

  1. Rationales Are Not Silver Bullets: Measuring the Impact of Rationales on Model Performance and Reliability

    cs.CL 2025-05 conditional novelty 7.0 of 10

    Rationale-augmented finetuning can hurt accuracy while improving calibration, with the sizes of both effects tied linearly to task difficulty.

  2. From Legal Text to Tech Specs: Generative AI's Interpretation of Consent in Privacy Law

    cs.SE 2025-07 conditional novelty 4.0 of 10

    An LLM pipeline that flags and fixes non-compliant consent use cases works imperfectly: it catches about two-thirds of relevant cases with reasoning prompts, and most of its fixes are legally sound but logically inconsistent.

Pith tools