Skip to content
← Back to explorer

Learning to Answer from Correct Demonstrations

Nirmit Joshi, Gene Li, Siddharth Bhandari, Shiva Prasad Kasiviswanathan, Cong Ma, Nathan Srebro · Oct 17, 2025 · Citations: 0

Abstract

We study the problem of learning to generate an answer (or completion) to a question (or prompt), where there could be multiple correct answers, any one of which is acceptable at test time. Learning is based on demonstrations of some correct answer to each training question, as in Supervised Fine Tuning (SFT). We formalize the problem as imitation learning (i.e., apprenticeship learning) in contextual bandits, with offline demonstrations from some expert (optimal, or very good) policy, without explicitly observed rewards. In contrast to prior work, which assumes the demonstrator belongs to a bounded-complexity policy class, we propose relying only on the underlying reward model (i.e., specifying which answers are correct) being in a bounded-complexity class, which we argue is a strictly weaker assumption. We show that likelihood-maximization methods can fail in this setting, and instead present an approach that learns to answer nearly as well as the demonstrator, with sample complexity logarithmic in the cardinality of the reward class. Our method is similar to Syed and Schapire 2007, when adapted to a contextual bandit (i.e., single step) setup, but is a simple one-pass online approach that enjoys an "optimistic rate" (i.e., $1/\varepsilon$ when the demonstrator is optimal, versus $1/\varepsilon^2$ in Syed and Schapire), and works even with arbitrarily adaptive demonstrations.

HFEPX Relevance Assessment

This paper has direct human-feedback and/or evaluation protocol signal and is likely useful for eval pipeline design.

Eval-Fit Score

55/100 • Medium

Useful as a secondary reference; validate protocol details against neighboring papers.

Human Feedback Signal

Detected

Evaluation Signal

Detected

HFEPX Fit

High-confidence candidate

Human Data Lens

  • Uses human feedback: Yes
  • Feedback types: Demonstrations
  • Rater population: Domain Experts
  • Unit of annotation: Unknown
  • Expertise required: General
  • Extraction source: Persisted extraction

Evaluation Lens

  • Evaluation modes: Automatic Metrics
  • Agentic eval: None
  • Quality controls: Not reported
  • Confidence: 0.65
  • Flags: None

Protocol And Measurement Signals

Benchmarks / Datasets

No benchmark or dataset names were extracted from the available abstract.

Reported Metrics

No metric terms were extracted from the available abstract.

Research Brief

Deterministic synthesis

We study the problem of learning to generate an answer (or completion) to a question (or prompt), where there could be multiple correct answers, any one of which is acceptable at test time. HFEPX signals include Demonstrations, Automatic Metrics with confidence 0.65. Updated from current HFEPX corpus.

Generated Mar 3, 2026, 4:42 PM · Grounded in abstract + metadata only

Key Takeaways

  • We study the problem of learning to generate an answer (or completion) to a question (or prompt), where there could be multiple correct answers, any one of which is acceptable at…
  • Learning is based on demonstrations of some correct answer to each training question, as in Supervised Fine Tuning (SFT).
  • Primary extracted protocol signals: Demonstrations, Automatic Metrics.

Researcher Actions

  • Compare its human-feedback setup against pairwise and rubric hubs.
  • Identify benchmark choices from full text before operationalizing conclusions.
  • Verify metric definitions before comparing against your eval pipeline.

Caveats

  • Generated from title, abstract, and extracted metadata only; full-paper implementation details are not parsed.
  • Extraction confidence is probabilistic and should be validated for critical decisions.

Research Summary

Contribution Summary

  • We study the problem of learning to generate an answer (or completion) to a question (or prompt), where there could be multiple correct answers, any one of which is acceptable at test time.
  • Learning is based on demonstrations of some correct answer to each training question, as in Supervised Fine Tuning (SFT).
  • We formalize the problem as imitation learning (i.e., apprenticeship learning) in contextual bandits, with offline demonstrations from some expert (optimal, or very good) policy, without explicitly observed rewards.

Researcher Checklist

  • Pass: Human feedback protocol is explicit

    Detected: Demonstrations

  • Pass: Evaluation mode is explicit

    Detected: Automatic Metrics

  • Gap: Quality control reporting appears

    No calibration/adjudication/IAA control explicitly detected.

  • Gap: Benchmark or dataset anchors are present

    No benchmark/dataset anchor extracted from abstract.

  • Gap: Metric reporting is present

    No metric terms extracted.

Related Papers

Papers are ranked by protocol overlap, extraction signal alignment, and semantic proximity.

Need human evaluators for your AI research? Scale annotation with expert AI Trainers.