Article
CommunityJev Models & Reimplementations
ReListen: Training-Free Evidence Localization in Hour-Long Audio
arXiv paper introducing ReListen, a training-free method in which a frozen audio language model runs a Jev-style scan: it scores each 20-second window by the probability it helps answer a question, then answers from the highest-scoring minute of hour-long recordings.
- Format
- Article
- Reading time
- 2 min
- Published by
- Community
- Added
- Nov 2026
Highlights
- Training-free: a frozen audio language model finds its own evidence before answering.
- Scores each 20-second window by the probability it helps answer the question.
- Selected minute overlaps annotated evidence for 88.5% of VoiceGiraffe questions.
- Exceeds a whole-recording transcript by 9.2 points on VoiceGiraffe.
- Answers faster than every transcript-based system with constant memory.
Questions about this entry
Short answers drawn from the linked project or page.
A frozen audio language model scores each 20-second window by the probability that it helps answer the question, then answers from the highest-scoring minute.
It is the scoring pass in which the model assigns each 20-second window a probability that the window helps answer the question, before answering from the top-scoring minute.
For 88.5% of VoiceGiraffe questions, the highest-scoring minute overlaps the annotated evidence, more often than transcript retrieval.
No. The method is training-free and works directly on audio, and the paper reports it answers faster than every transcript-based system.
It matches or exceeds hearing the whole recording using a small fraction of the audio, and on VoiceGiraffe it exceeds a transcript of the whole recording by 9.2 points.
Given the same excerpts, reading their transcript matches listening on speech content, but listening is clearly more accurate on questions about sounds and voices.
Its memory stays constant even for recordings too long to hear whole on one GPU, and it grounds events on LAT-Bench as well as a model trained for temporal grounding.
More Jev resources like this
Open-weight vision-language judge that takes a task instruction plus 10 frames of a robot manipulation attempt and predicts success or failure with calibrated probabilities in a single forward pass. Released with inference code and a 20,628-prediction evaluation set.
Jev Models & ReimplementationsArticle#arxiv#open-models#evaluation
Community
arXiv paper proposing JevRec, which encodes recommendation items as variable-length semantic identifiers and adds a Jev-like decision modeling component with separate heads for length prediction and content generation, evaluated on three public datasets.
Jev Models & ReimplementationsArticle#arxiv#research#benchmark
Community
arXiv paper on a small candidate formalizer trained on offline teacher-generated lists that builds explicit, validated alternatives from task context for a frozen decision model, evaluated on CALVIN-D episodes and annotations.
Jev Models & ReimplementationsArticle#arxiv#research#benchmark
Community
A family of 0.8B-9B ophthalmic decision models that score candidate answers without autoregressive decoding, trained on RetiJev-Data: 78,158 clinical-decision and diagnosis questions and 50,074 guideline-grading questions. RetiJev-9B answers a case in 158 ms on one GPU.
Jev Models & ReimplementationsArticle#arxiv#open-models#benchmark
Community
arXiv paper on Decision-Level Continual Learning from Typed Feedback, a method that keeps Jev-style accept and revoke signals in bounded channels above a frozen scorer, and introduces TCPA-Bench, which credits a correction only through its effect on later decisions.
Jev Models & ReimplementationsArticle#arxiv#research#evaluation
Community
Fine-tuning method that turns fix commits into comparison groups so an open-weight model scores code functions for vulnerability. An ensemble of five Qwen3.5-4B models reaches 0.814 AUROC on a 7,270-function benchmark, against 0.743 for Jev 1.13.
Jev Models & ReimplementationsArticle#arxiv#open-models#security
Community
From the Jev guides
Original write-ups on what this entry is for.
From the Jev AI model community
Posts from builders shipping with Jev right now.