Papers › Grounding Language Models for Visual Entity Recognition

Grounding Language Models for Visual Entity Recognition

28 Feb 2024arXiv:2402.18695archive 2025-07-28

Zilin Xiao, Ming Gong, Paola Cascante-Bonilla, Xingyao Zhang, Jie Wu, Vicente Ordonez

We introduce AutoVER, an Autoregressive model for Visual Entity Recognition. Our model extends an autoregressive Multi-modal Large Language Model by employing retrieval augmented constrained generation. It mitigates low performance on out-of-domain entities while excelling in queries that require visually-situated reasoning. Our method learns to distinguish similar entities within a vast label space by contrastively training on hard negative pairs in parallel with a sequence-to-sequence objective without an external retriever. During inference, a list of retrieved candidate answers explicitly guides language generation by removing invalid decoding paths. The proposed method achieves significant improvements across different dataset splits in the recently proposed Oven-Wiki benchmark. Accuracy on the Entity seen split rises from 32.7% to 61.5%. It also demonstrates superior performance on the unseen and query splits by a substantial double-digit margin.

PaperPDFCodeCode Syntology ranMethod as code

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

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

For agents, on Syntology's MCP service (how to connect):

Code

Syntology 26 code samples harvested from 1 repository linked to this paper: 12 ran (of which 2 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 0 violated, 2 with no contract checked; 10 where Syntology's instrument failed) · 14 unverified. Where Syntology's instrument failed: 8 where our contract draft was wrong; 2 where the contract check could not run; failures of Syntology's instrument, not of the code.

By repository: official repository: 26 samples from 1 repository, 12 ran (2 with no instrument failure; 10 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

Levels written between 2026-09-22 and 2026-09-24 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

mrzilinxiao/autover officialmentioned in papermentioned on GitHubpytorchApache-2.0Syntology: 26 samples, 12 ran (10 instrument-failed) · 14 unverified 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

26 samples harvested: 12 ran (of which 2 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 0 violated, 2 with no contract checked; 10 where Syntology's instrument failed) · 14 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

8ran · our draft was wrong
2ran; the contract check could not run
2ran
14unverified

Licence: 1 of the 26 samples is pointer only, meaning Syntology does not serve that copy's text. For that sample, this paper's copy is under a permissive licence and cleared, but the stored copy of the same code that Syntology serves is not; the licence label's tooltip gives Syntology's recorded reason. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from mrzilinxiao/autover. “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. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (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 are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. 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 six 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; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. 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.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity. A census tag appears when Syntology's census ran implementations with that name on one shared input per bucket: it gives how many were compared, from how many papers, and how many distinct outputs they produced, and names this sample's bucket when this sample was one of them.

mrzilinxiao/autover official repository 26 samples, 12 ran (10 instrument-failed) · 14 unverified
CLIPVisionTower in 11 papers LLaVA/llava/model/language_model/llava_llama.py ran Apache-2.0 (permissive) · c41616ca93fd79a8 · report fixture: returned CLIPVisionTower · level recorded 2026-09-22
CausalLMOutputWithPastWithRet LLaVA/llava/model/language_model/llava_llama.py ran Apache-2.0 (permissive) · d3336c4c93426ff0 · report fixture: returned CausalLMOutputWithPastWithRet · level recorded 2026-09-22
accuracy in 872 papers
census: 8 implementations from 7 papers, 7 distinct outputs across 6 buckets, one shared input per bucket
LLaVA/llava/model/language_model/llava_llama.py ran; the contract check could not run (RAISES) Apache-2.0 (permissive) · ed5ed903e9e24d08 · report fixture: signature output:Tensor/2/float, target:Tensor/1/int, padding:int/0/int, topk:tuple/1/int · returned list · level recorded 2026-09-22
contrastive_acc LLaVA/llava/model/language_model/llava_llama.py ran; the contract check could not run (RAISES) Apache-2.0 (permissive) · 0b76a505072cf34e · report fixture: signature logits:Tensor/2/float, target:Tensor/2/int, topk:tuple/1/int · returned list · level recorded 2026-09-22
contrastive_loss in 80 papers
census: 11 implementations from 18 papers, 7 distinct outputs across 3 buckets, one shared input per bucket this sample: bucket 1 of 3, arg 1: rank 2, kind float, dtype float32
LLaVA/llava/model/language_model/llava_llama.py ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · 83ada93fd84968cc · report fixture: signature logits:Tensor/2/float · returned Tensor · level recorded 2026-09-22
contrastive_loss_with_target common_utils/loss_utils.py ran · our draft was wrong Apache-2.0 (permissive) · 418757e4ade61509 · report fixture: signature logits:Tensor/2/float, target:Tensor/1/int · returned Tensor · level recorded 2026-09-24
encode_image in 131 papers baselines/submit_gpt4v.py ran · our draft was wrong Apache-2.0 (permissive) here; the stored copy is pointer only · f41cb1a19b154297 · report fixture: signature image_path:str/0/str · returned str · level recorded 2026-09-22
evaluate_oven bm25_toolkit/official_oven_eval_toolkit.py ran · our draft was wrong Apache-2.0 (permissive) · 249898a5ceca22bd · report fixture: signature ref:list/1/dict, pred:list/1/dict · returned dict · level recorded 2026-09-24
load_json in 344 papers bm25_toolkit/bm25_index.py ran · our draft was wrong Apache-2.0 (permissive) · e18691c020115ad3 · report fixture: signature path:str/0/str · returned list · level recorded 2026-09-22
load_jsonl in 179 papers bm25_toolkit/official_oven_eval_toolkit.py ran · our draft was wrong Apache-2.0 (permissive) · 0dc083025b2aad97 · report fixture: signature path:str/0/str · returned list · level recorded 2026-09-24
prepare_qid2example bm25_toolkit/official_oven_eval_toolkit.py ran · our draft was wrong Apache-2.0 (permissive) · cb38c498b185418c · report fixture: signature reference:list/1/dict · returned dict · level recorded 2026-09-24
run_exactmatch_query bm25_toolkit/exactmatch_query.py ran · our draft was wrong Apache-2.0 (permissive) · 0ac33fc045371be4 · report fixture: signature query:dict/0/? · returned dict · level recorded 2026-09-24
LlavaLlamaForCausalLMWithRet LLaVA/llava/model/language_model/llava_llama.py unverified Apache-2.0 (permissive) · 8db3d394148aae40 · report
LlavaLlamaModelWithRet LLaVA/llava/model/language_model/llava_llama.py unverified Apache-2.0 (permissive) · bd94789c33b6b3ce · report
LlavaMetaForCausalLMWithRet LLaVA/llava/model/language_model/llava_llama.py unverified Apache-2.0 (permissive) · 3653c732c4369b00 · report
LlavaMetaModel in 16 papers LLaVA/llava/model/language_model/llava_llama.py unverified Apache-2.0 (permissive) · ee2ab77aead57d40 · report
RetrieverMetaModel LLaVA/llava/model/language_model/llava_llama.py unverified Apache-2.0 (permissive) · 31d0e17da04e4bfe · report
all_gather_with_grad in 93 papers common_utils/dist_utils.py unverified Apache-2.0 (permissive) · 0ec9fc2025c16f65 · report
all_gather_with_grad_torch common_utils/dist_utils.py unverified Apache-2.0 (permissive) · f1a8171c466e7574 · report
build_entity_encoder LLaVA/llava/model/language_model/llava_llama.py unverified Apache-2.0 (permissive) · e374a58821b3513a · report
build_vision_tower in 35 papers LLaVA/llava/model/language_model/llava_llama.py unverified Apache-2.0 (permissive) · 01a59cbdcf1329ec · report
find_img_path baselines/submit_gpt4v.py unverified Apache-2.0 (permissive) · e8003b1ffefe6856 · report
get_peft_state_maybe_zero_3 in 31 papers train_oven.py unverified Apache-2.0 (permissive) · fa1225dfac92bc0d · report
get_peft_state_non_lora_maybe_zero_3 in 24 papers train_oven.py unverified Apache-2.0 (permissive) · 1c53657305b66e9f · report
maybe_zero_3 in 74 papers train_oven.py unverified Apache-2.0 (permissive) · 616ffbdc154ed2d8 · report
run_gpt_query bm25_toolkit/bm25_query.py unverified Apache-2.0 (permissive) · a5be46c82b7c5a66 · report

Method as code Syntology

What Syntology's graph records as this paper's claimed method, lifted into code: each assembly names its bricks (functions and classes), each with its own role and its own verification level. An assembly is not a check that the code implements the method correctly or reproduces the paper's results. Levels are each brick's own verification level in Syntology's graph (level 0: no recorded run), listed brick by brick and never combined into one level; “ran” means the brick executed on a synthesized input, not that its output is correct, and the instrument figure counts failures of Syntology's instrument, not of the code. Where a model read the code against the paper's claim, each assembly's label verdict is one model's judgement, kept in this page's JSON twin; it is not calibrated, so no card prints it.

1 assembly on this paper: 1 harvested, 0 generated.

harvested Assembly for AutoVER (graph method: AutoVER). 17 bricks; individually on a fixture: 9 ran (of which 5 constructed an object rather than computing a result; 5 with no instrument failure: 0 honoured, 0 violated, 5 with no contract checked; 4 where Syntology's instrument failed) · 8 unverified. The assembly has not been run end-to-end.

  1. LlavaLlamaForCausalLMWithRet entry · level 0 · unverified · 8db3d394148aae40
  2. CausalLMOutputWithPastWithRet model · level 1 · ran · d3336c4c93426ff0
  3. LlavaConfig helper · level 1 · ran · e9cfb7ae423d3e3a
  4. LlavaLlamaModelWithRet model · level 0 · unverified · bd94789c33b6b3ce
  5. LlavaMetaForCausalLM helper · level 0 · unverified · abf7b5851a13bc76
  6. LlavaMetaForCausalLMWithRet helper · level 0 · unverified · 3653c732c4369b00
  7. contrastive_acc helper · level 1 · ran; the contract check could not run (RAISES) · 0b76a505072cf34e
  8. contrastive_loss helper · level 1 · ran · our draft was wrong · 83ada93fd84968cc
  9. LlavaMetaModel helper · level 0 · unverified · ee2ab77aead57d40
  10. RetrieverMetaModel helper · level 0 · unverified · 31d0e17da04e4bfe
  11. accuracy helper · level 1 · ran; the contract check could not run (RAISES) · ed5ed903e9e24d08
  12. build_vision_projector helper · level 1 · ran · our draft was wrong · ef4db0da176c674c
  13. build_vision_tower helper · level 0 · unverified · 01a59cbdcf1329ec
  14. build_entity_encoder helper · level 0 · unverified · e374a58821b3513a
  15. IdentityMap helper · level 1 · ran · 9f0f5d62445adabe
  16. CLIPVisionTower helper · level 1 · ran · c41616ca93fd79a8
  17. GatherLayer helper · level 1 · ran · a839c0c942f1b694

Each brick ends with its code_sha256, Syntology's identity for that exact code; click one to copy the get_code call that returns it. For agents, on Syntology's MCP service: assembly(arxiv_id="2402.18695") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

Language ModelingLanguage ModellingLarge Language ModelRetrievalText Generation

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

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