Papers › Generate, but Verify: Reducing Hallucination in Vision-Language Models with...

Generate, but Verify: Reducing Hallucination in Vision-Language Models with Retrospective Resampling

17 Apr 2025arXiv:2504.13169archive 2025-07-28

Tsung-Han Wu, HeeKyung Lee, Jiaxin Ge, Joseph E. Gonzalez, Trevor Darrell, David M. Chan

Vision-Language Models (VLMs) excel at visual understanding but often suffer from visual hallucinations, where they generate descriptions of nonexistent objects, actions, or concepts, posing significant risks in safety-critical applications. Existing hallucination mitigation methods typically follow one of two paradigms: generation adjustment, which modifies decoding behavior to align text with visual inputs, and post-hoc verification, where external models assess and correct outputs. While effective, generation adjustment methods often rely on heuristics and lack correction mechanisms, while post-hoc verification is complicated, typically requiring multiple models and tending to reject outputs rather than refine them. In this work, we introduce REVERSE, a unified framework that integrates hallucination-aware training with on-the-fly self-verification. By leveraging a new hallucination-verification dataset containing over 1.3M semi-synthetic samples, along with a novel inference-time retrospective resampling technique, our approach enables VLMs to both detect hallucinations during generation and dynamically revise those hallucinations. Our evaluations show that REVERSE achieves state-of-the-art hallucination reduction, outperforming the best existing methods by up to 12% on CHAIR-MSCOCO and 28% on HaloQuest. Our dataset, model, and code are available at: https://reverse-vlm.github.io.

PaperPDFCodeCode Syntology ran

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, Syntology's MCP tool lists every function and class Syntology harvested from this paper and whether it ran (how to connect): get_harvested_code_for_paper(arxiv_id="2504.13169")

Code

Syntology Ran 10 of 11 code samples harvested from 1 repository linked to this paper; 1 has no recorded run. Of those that ran: 1 ran · violated contract; 6 ran · our draft was wrong; 3 ran · fixture could not drive it.

By repository: official repository: 9 samples from 1 repository, 8 ran; 2 identical to code first harvested elsewhere. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

tsunghan-wu/reverse_vlm officialmentioned in papermentioned on GitHubpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

11 samples harvested; 10 ran; 0 honoured the contract we drafted; 1 has no recorded run. Read from Syntology's graph 2026-09-24; that is when this build read the record, not when the samples ran.

1ran · violated contract
6ran · our draft was wrong
3ran · fixture could not drive it
1unverified

Licence: 11 of the 11 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from tsunghan-wu/reverse_vlm. Some samples are identical code Syntology first harvested from another repository; for those, this paper's copy is not located and its licence is not recorded. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself; “our draft was wrong” and “fixture could not drive it” are failures of Syntology's instrument, not of the code.

Each sample ends with its code_sha256, Syntology's identity for that exact code. An agent fetches the stored sample with Syntology's MCP tool get_code(code_sha256="…") (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples from a repository marked official are listed first. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of four reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; or the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar). Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

back_track_last_sentence tsunghan-wu/reverse_vlm/llava/eval/inference_utils.py official repository ran · our draft was wrong no licence file found · pointer only · 3a19bf415091befc · report
check_special_token_mismatch tsunghan-wu/reverse_vlm/llava/eval/inference_utils.py official repository ran · violated contract fingerprinted no licence file found · pointer only · 194dfaa5e36a26ea · report
combine_coco_instances tsunghan-wu/reverse_vlm/scripts/eval/eval_chair.py official repository ran · our draft was wrong no licence file found · pointer only · 66415dc1dd50f555 · report
de_hallucination_postprocessing tsunghan-wu/reverse_vlm/llava/eval/inference_utils.py official repository ran · fixture could not drive it no licence file found · pointer only · 527ddfe8465dda89 · report
do_inference tsunghan-wu/reverse_vlm/llava/eval/inference_utils.py official repository ran · our draft was wrong no licence file found · pointer only · f13b7a81d4dfe183 · report
early_exit tsunghan-wu/reverse_vlm/llava/eval/inference_utils.py official repository ran · our draft was wrong no licence file found · pointer only · f7756cda9b536a1b · report
load_generated_captions tsunghan-wu/reverse_vlm/scripts/eval/eval_chair.py official repository ran · our draft was wrong no licence file found · pointer only · 897c16b1c81103f4 · report
render_output tsunghan-wu/reverse_vlm/llava/eval/inference_utils.py official repository ran · fixture could not drive it no licence file found · pointer only · ee5e60b564c5eff5 · report
retrospective_resampling_generate tsunghan-wu/reverse_vlm/llava/eval/inference_utils.py official repository unverified no licence file found · pointer only · f8f2edbc52243524 · report
combine_coco_captions identical code first harvested elsewhere ran · our draft was wrong licence of this copy not recorded · 25585704d77cafb4 · report
unpad_image identical code first harvested elsewhere ran · fixture could not drive it fingerprinted licence of this copy not recorded · 55c32993da87759b · report

Tasks

Hallucination

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

ALIGN

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections