Pith. sign in

REVIEW 8 cited by

Language Model Cascades

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 2207.10342 v2 pith:OZQZLIHB submitted 2022-07-21 cs.CL cs.AI

classification cs.CLcs.AI
keywords languagemodelmodelscascadesprobabilistictechniquesabilitiesallow
verification ladder T0 review T1 audit T2 compute T3 formal

Signed reviews

No signed human review yet.

0 comments
read the original abstract

Prompted models have demonstrated impressive few-shot learning abilities. Repeated interactions at test-time with a single model, or the composition of multiple models together, further expands capabilities. These compositions are probabilistic models, and may be expressed in the language of graphical models with random variables whose values are complex data types such as strings. Cases with control flow and dynamic structure require techniques from probabilistic programming, which allow implementing disparate model structures and inference strategies in a unified language. We formalize several existing techniques from this perspective, including scratchpads / chain of thought, verifiers, STaR, selection-inference, and tool use. We refer to the resulting programs as language model cascades.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 8 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. OpenAlex reports about 22 citations worldwide. Full citation record

  1. Why Ask One When You Can Ask $k$? Learning-to-Defer to the Top-$k$ Experts

    cs.LG 2025-04 unverdicted novelty 8.0 of 10

    This paper introduces top-k learning-to-defer, where a single k-independent policy routes each query to the k highest-scored experts, plus an adaptive variant that chooses k per query.

  2. Auto: The AGI Compiler

    cs.LG 2026-07 conditional novelty 7.0 of 10

    AUTO compiles witnessed-deterministic LLM-agent spans into verified WASM cognition binaries and recompiles on deopt, cutting cost 6.4× at 96.9% parity on a 300-item shifted stream.

  3. Conformal Cascade: Distribution-Free Accuracy Guarantees for Multi-Tier LLM Inference

    cs.LG 2026-07 conditional novelty 6.0 of 10

    Using conformal prediction-set size as the cascade deferral rule yields distribution-free cascade accuracy bounds and usually beats confidence-threshold heuristics on multiple-choice LLM benchmarks.

  4. Modeling Open-World Cognition as On-Demand Synthesis of Probabilistic Models

    cs.CL 2025-07 conditional novelty 6.0 of 10

    A hybrid language-model and probabilistic-program architecture predicts human judgments on novel open-world reasoning vignettes better than language-model-only baselines.

  5. Understanding the Logic of Direct Preference Alignment through Logic

    cs.CL 2024-12 conditional novelty 6.0 of 10

    Direct preference alignment losses can be expressed as logical programs over model predictions, yielding an organized landscape of billions of definable losses and a route to new variants.

  6. What makes prompts a graph: necessary and sufficient conditions for prompt graph engineering

    cs.AI 2026-07 conditional novelty 5.0 of 10

    The paper defines prompt graph engineering via four necessary and sufficient conditions (explicit structure, structure/content separation, executable semantics, first-class artifact) and an inclusion/exclusion test th...

  7. Template-Based Financial Report Generation in Agentic and Decomposed Information Retrieval

    cs.IR 2025-04 conditional novelty 5.0 of 10

    Decomposing a report template into per-section subqueries yields more detailed and better-scored financial reports than a multi-agent system that uses the full template at once.

  8. MultiQ&A: An Analysis in Measuring Robustness via Automated Crowdsourcing of Question Perturbations and Answers

    cs.CL 2025-02 conditional novelty 4.0 of 10

    MultiQ&A automatically rephrases questions and crowdsources answers from LLM agents, showing gpt-3.5-turbo is fairly stable on extractive and multiple-choice QA but more variable on free-form answers.

Pith tools