MétaCan
Menu
Back to cohort

Can Retriever-Augmented Language Models Reason? The Blame Game Between the Retriever and the Language Model

2023· article· en· W4389524581 on OpenAlexaffabout
Parishad BehnamGhader, Santiago Miret, Siva Reddy

Bibliographic record

Venuenot available
Typearticle
Languageen
FieldComputer Science
TopicTopic Modeling
Canadian institutionsMcGill University
Fundersnot available
KeywordsComputer scienceLanguage modelNatural language processingArtificial intelligenceMetric (unit)

Abstract

fetched live from OpenAlex

Augmenting pretrained language models with retrievers has shown promise in effectively solving common NLP problems, such as language modeling and question answering.In this paper, we evaluate the strengths and weaknesses of popular retriever-augmented language models, namely kNN-LM, REALM, DPR + FiD, Contriever + ATLAS, and Contriever + Flan-T5, in reasoning over retrieved statements across different tasks.Our findings indicate that the simple similarity metric employed by retrievers is insufficient for retrieving all the necessary statements for reasoning.Additionally, the language models do not exhibit strong reasoning even when provided with only the required statements.Furthermore, when combined with imperfect retrievers, the performance of the language models becomes even worse, e.g., Flan-T5's performance drops by 28.6% when retrieving 5 statements using Contriever.While larger language models improve performance, there is still a substantial room for enhancement.Our further analysis indicates that multihop retrieveand-read is promising for large language models like GPT-3.5, but does not generalize to other language models like Flan-T5-xxl. 1

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

How this classification was reachedexpand

Full frame machine prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.

metaresearch head score (Codex)0.007
metaresearch head score (Gemma)0.029
Version: metacan-v3-hybrid-931329e0061cValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Simulation or modeling · Consensus signal: none
GenreCandidate signal: Empirical · Consensus signal: none
Teacher disagreement score0.011
Threshold uncertainty score0.037

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.0070.029
Meta-epidemiology (narrow)0.0020.001
Meta-epidemiology (broad)0.0010.003
Bibliometrics0.0010.001
Science and technology studies0.0010.002
Scholarly communication0.0040.017
Open science0.0040.004
Research integrity0.0020.005
Insufficient payload (model declined to judge)0.0090.006

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.023
GPT teacher head0.253
Teacher spread0.229 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.

The models applied no category: nothing in the taxonomy fit this work.
Study designSimulation or modeling
Domainnot available
GenreEmpirical

How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".

Quick stats

Citations8
Published2023
Admission routes2
Has abstractyes

Explore more

Same topicTopic ModelingFrench-language works237,207