Paper Citation Record · LEDGER

Beyond Global Editing: Per-Instance Disentangled Subspaces for Training-Free Hallucination Mitigation in LVLMs

As of 17 August 2026, this Paper Citation Record lists 38 of 38 outbound references and 0 inbound Pith citation observations for arXiv:2608.09344.

A citation records a reference. It does not transfer a finding from one paper to another.

pith.paper-citation-record.v1
2608.09344 v1

Coverage vector

measured 38 of 38 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-11T19:01:40.526644Z

measured 38 of 38 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-17T06:30:58.91139+00:00

measured 0 of 0 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links

measured 0 of 1 external citation measurements

A source-named dated measurement, never combined with another source.

Source: cited_works

Reference resolution

38 of 38 outbound references displayed

verified exact1
verified fuzzy17
unresolved19
parse uncertain0
malformed identifier1
metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation f71fc931-a6ac-4e22-8230-a0c1a09e683f · outbound

This paper cites Foundation model-powered 3d few- shot class incremental learning via training-free adaptor.

Beyond Global Editing: Per-Instance Disentangled Subspaces for Training-Free Hallucination Mitigation in LVLMs Foundation model-powered 3d few- shot class incremental learning via training-free adaptor

Reference 1

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T19:02:26.595872Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 0dceb04f-f9b6-4213-9eb0-835f3558e522 · outbound

This paper cites Hime: Mitigating object hallucinations in lvlms via hallucination in- sensitivity model editing, 2026.

Beyond Global Editing: Per-Instance Disentangled Subspaces for Training-Free Hallucination Mitigation in LVLMs Hime: Mitigating object hallucinations in lvlms via hallucination in- sensitivity model editing, 2026

Reference 2

Resolution: verified exact; raw_fallback, observed 2026-08-11T19:02:26.065746Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 236d5258-1cf9-489a-82b9-1c7db8cb93e8 · outbound

This paper cites Qwen-VL: A Versatile Vision-Language Model for Understanding, Localization, Text Reading, and Beyond.

Beyond Global Editing: Per-Instance Disentangled Subspaces for Training-Free Hallucination Mitigation in LVLMs Qwen-VL: A Versatile Vision-Language Model for Understanding, Localization, Text Reading, and Beyond

Reference 3

Resolution: unresolved; no resolver link, observed 2026-08-11T19:01:40.324352Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 648bfefd-3b2b-42fd-844e-6da1dcbabbbd · outbound

This paper cites HALC: Object Hallucination Reduction via Adaptive Focal-Contrast Decoding.

Beyond Global Editing: Per-Instance Disentangled Subspaces for Training-Free Hallucination Mitigation in LVLMs HALC: Object Hallucination Reduction via Adaptive Focal-Contrast Decoding

Reference 4

Resolution: unresolved; no resolver link, observed 2026-08-11T19:01:40.330288Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d6b83965-ff75-462f-9f5c-36e982adf29b · outbound

This paper cites Synthesized feature based few-shot class-incremental learning on a mixture of subspaces.

Beyond Global Editing: Per-Instance Disentangled Subspaces for Training-Free Hallucination Mitigation in LVLMs Synthesized feature based few-shot class-incremental learning on a mixture of subspaces

Reference 5

Resolution: unresolved; no resolver link, observed 2026-08-11T19:01:40.335892Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 888e6b34-089e-4f81-808f-4b467d730af6 · outbound

This paper cites Canonical shape projection is all you need for 3d few-shot class incremental learning.

Beyond Global Editing: Per-Instance Disentangled Subspaces for Training-Free Hallucination Mitigation in LVLMs Canonical shape projection is all you need for 3d few-shot class incremental learning

Reference 6

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T19:02:26.568726Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 6fa7db8e-fdb1-4862-bf29-18855eeb133f · outbound

This paper cites DoLa: Decoding by Contrasting Layers Improves Factuality in Large Language Models.

Beyond Global Editing: Per-Instance Disentangled Subspaces for Training-Free Hallucination Mitigation in LVLMs DoLa: Decoding by Contrasting Layers Improves Factuality in Large Language Models

Reference 7

Resolution: unresolved; no resolver link, observed 2026-08-11T19:01:40.347941Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 872888cc-758d-4a74-b7a3-59638ae0b34e · outbound

This paper cites Glass, and Pengcheng He.

Beyond Global Editing: Per-Instance Disentangled Subspaces for Training-Free Hallucination Mitigation in LVLMs Glass, and Pengcheng He

Reference 8

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T19:02:26.539731Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 264c4d4c-ce8b-4821-8d16-731856f7fa59 · outbound

This paper cites Instructblip: Towards general-purpose vision-language models with instruction tuning, 2023.

Beyond Global Editing: Per-Instance Disentangled Subspaces for Training-Free Hallucination Mitigation in LVLMs Instructblip: Towards general-purpose vision-language models with instruction tuning, 2023

Reference 9

Resolution: unresolved; no resolver link, observed 2026-08-11T19:01:40.364750Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2e89a958-2f42-4629-8b37-ad8db3ba9d10 · outbound

This paper cites Etta: Efficient test-time adaptation for vision-language models through dynamic embedding updates.

Beyond Global Editing: Per-Instance Disentangled Subspaces for Training-Free Hallucination Mitigation in LVLMs Etta: Efficient test-time adaptation for vision-language models through dynamic embedding updates

Reference 10

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T19:02:26.482461Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 73be7c55-c081-4e0b-80e1-ad04e4584f5e · outbound

This paper cites Test-time adaptation of 3d point clouds via denoising diffusion models.

Beyond Global Editing: Per-Instance Disentangled Subspaces for Training-Free Hallucination Mitigation in LVLMs Test-time adaptation of 3d point clouds via denoising diffusion models

Reference 11

Resolution: malformed identifier; no resolver link, observed 2026-08-11T19:01:40.381735Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8587b63e-a23b-485d-b003-6b128cf96f1f · outbound

This paper cites Fighting hallucinations with counterfactuals: Diffusion-guided perturbations for lvlm halluci- nation suppression.

Beyond Global Editing: Per-Instance Disentangled Subspaces for Training-Free Hallucination Mitigation in LVLMs Fighting hallucinations with counterfactuals: Diffusion-guided perturbations for lvlm halluci- nation suppression

Reference 12

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T19:02:26.437518Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 84f22983-704c-4e2b-a32c-88466cb7f4d0 · outbound

This paper cites An image is worth 16x16 words: Transformers for image recognition at scale.

Beyond Global Editing: Per-Instance Disentangled Subspaces for Training-Free Hallucination Mitigation in LVLMs An image is worth 16x16 words: Transformers for image recognition at scale

Reference 13

Resolution: unresolved; no resolver link, observed 2026-08-11T19:01:40.392147Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9e56e670-ce02-485b-ac55-e663b1e5ce5a · outbound

This paper cites Beam Search Strategies for Neural Machine Translation.

Beyond Global Editing: Per-Instance Disentangled Subspaces for Training-Free Hallucination Mitigation in LVLMs Beam Search Strategies for Neural Machine Translation

Reference 14

Resolution: unresolved; no resolver link, observed 2026-08-11T19:01:40.397256Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5f4bcca5-61d9-4d50-977f-355f0910554c · outbound

This paper cites Opera: Alleviating hallucination in multi-modal large language models via over-trust penalty and retrospection-allocation.

Beyond Global Editing: Per-Instance Disentangled Subspaces for Training-Free Hallucination Mitigation in LVLMs Opera: Alleviating hallucination in multi-modal large language models via over-trust penalty and retrospection-allocation

Reference 15

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T19:02:26.402521Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation c2f2abce-dc9f-40cb-a4e9-2cc889b2983e · outbound

This paper cites Mitigating object hallucinations in large vision-language models through visual contrastive decoding.

Beyond Global Editing: Per-Instance Disentangled Subspaces for Training-Free Hallucination Mitigation in LVLMs Mitigating object hallucinations in large vision-language models through visual contrastive decoding

Reference 17

Resolution: unresolved; no resolver link, observed 2026-08-11T19:01:40.412027Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3f521cb8-507a-4205-a827-b9c87bd707d1 · outbound

This paper cites Blip-2: bootstrapping language-image pre-training with frozen image encoders and large language models.

Beyond Global Editing: Per-Instance Disentangled Subspaces for Training-Free Hallucination Mitigation in LVLMs Blip-2: bootstrapping language-image pre-training with frozen image encoders and large language models

Reference 18

Resolution: unresolved; no resolver link, observed 2026-08-11T19:01:40.416299Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 726235b3-3655-4407-a98a-4af3801e1bff · outbound

This paper cites Blip-2: Bootstrapping language-image pre-training with frozen image encoders and large language models.

Beyond Global Editing: Per-Instance Disentangled Subspaces for Training-Free Hallucination Mitigation in LVLMs Blip-2: Bootstrapping language-image pre-training with frozen image encoders and large language models

Reference 19

Resolution: unresolved; no resolver link, observed 2026-08-11T19:01:40.421470Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6cf65e59-ab1a-4248-aabd-4b89a57076eb · outbound

This paper cites Evaluating Object Hallucination in Large Vision-Language Models.

Beyond Global Editing: Per-Instance Disentangled Subspaces for Training-Free Hallucination Mitigation in LVLMs Evaluating Object Hallucination in Large Vision-Language Models

Reference 20

Resolution: unresolved; no resolver link, observed 2026-08-11T19:01:40.426471Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 11e7887a-c32c-4a0d-85f8-3784c84f85de · outbound

This paper cites Visual instruction tuning, 2023.

Beyond Global Editing: Per-Instance Disentangled Subspaces for Training-Free Hallucination Mitigation in LVLMs Visual instruction tuning, 2023

Reference 21

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T19:02:26.337959Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 437910c3-5229-45f0-bcfc-164a5f85188c · outbound

This paper cites Visual instruction tuning.

Beyond Global Editing: Per-Instance Disentangled Subspaces for Training-Free Hallucination Mitigation in LVLMs Visual instruction tuning

Reference 22

Resolution: unresolved; no resolver link, observed 2026-08-11T19:01:40.435546Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 14f90d9f-ec2d-4284-9a2e-61a8048239ea · outbound

This paper cites Reducing Hallucinations in Vision-Language Models via Latent Space Steering.

Beyond Global Editing: Per-Instance Disentangled Subspaces for Training-Free Hallucination Mitigation in LVLMs Reducing Hallucinations in Vision-Language Models via Latent Space Steering

Reference 23

Resolution: unresolved; no resolver link, observed 2026-08-11T19:01:40.441506Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a5cc5e7b-13f3-464b-9f96-d3be25a24c04 · outbound

This paper cites Learning transferable visual models from natural lan- guage supervision.

Beyond Global Editing: Per-Instance Disentangled Subspaces for Training-Free Hallucination Mitigation in LVLMs Learning transferable visual models from natural lan- guage supervision

Reference 24

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T19:02:26.295999Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation c0654ffb-06e1-47ee-86e3-c10492859174 · outbound

This paper cites Object Hallucination in Image Captioning.

Beyond Global Editing: Per-Instance Disentangled Subspaces for Training-Free Hallucination Mitigation in LVLMs Object Hallucination in Image Captioning

Reference 25

Resolution: unresolved; no resolver link, observed 2026-08-11T19:01:40.452003Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6b9be631-26a8-46dc-96eb-b3bdb4de3224 · outbound

This paper cites Aligning large multimodal models with factually augmented rlhf.

Beyond Global Editing: Per-Instance Disentangled Subspaces for Training-Free Hallucination Mitigation in LVLMs Aligning large multimodal models with factually augmented rlhf

Reference 26

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T19:02:26.270004Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 1cd2e62f-d671-4f05-95c2-00355208ae70 · outbound

This paper cites Mitigating hallucina- tions in large vision-language models with instruction contrastive decoding.

Beyond Global Editing: Per-Instance Disentangled Subspaces for Training-Free Hallucination Mitigation in LVLMs Mitigating hallucina- tions in large vision-language models with instruction contrastive decoding

Reference 27

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T19:02:26.242826Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 3efb084e-d4cf-4e96-b2b8-cf7dbd3514eb · outbound

This paper cites Detecting and mitigating hallucination in large vision language models via fine-grained ai feedback.

Beyond Global Editing: Per-Instance Disentangled Subspaces for Training-Free Hallucination Mitigation in LVLMs Detecting and mitigating hallucination in large vision language models via fine-grained ai feedback

Reference 28

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T19:02:26.197757Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 348f21f5-5a34-41c6-ab6a-dcf6e71bf216 · outbound

This paper cites Nullu: Mitigating object hallucinations in large vision-language models via halluspace projection.

Beyond Global Editing: Per-Instance Disentangled Subspaces for Training-Free Hallucination Mitigation in LVLMs Nullu: Mitigating object hallucinations in large vision-language models via halluspace projection

Reference 29

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T19:02:26.174627Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 55f0c92e-40ce-4d78-8d92-b431611aca3c · outbound

This paper cites an unresolved cited work.

Beyond Global Editing: Per-Instance Disentangled Subspaces for Training-Free Hallucination Mitigation in LVLMs Unresolved cited work

Reference 30

Resolution: unresolved; raw_fallback, observed 2026-08-11T19:02:26.218314Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 8077bbf1-fe5d-43f6-ba96-ef1e6f4e2add · outbound

This paper cites Woodpecker: Hallucination correction for multimodal large language models.Science China Information Sciences, 67(12): 220105, 2024.

Beyond Global Editing: Per-Instance Disentangled Subspaces for Training-Free Hallucination Mitigation in LVLMs Woodpecker: Hallucination correction for multimodal large language models.Science China Information Sciences, 67(12): 220105, 2024

Reference 31

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T19:02:26.132258Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 586061a6-830e-4686-97c9-ff8c8ee50e1e · outbound

This paper cites Rlhf-v: Towards trustworthy mllms via behavior alignment from fine-grained correctional human feedback.

Beyond Global Editing: Per-Instance Disentangled Subspaces for Training-Free Hallucination Mitigation in LVLMs Rlhf-v: Towards trustworthy mllms via behavior alignment from fine-grained correctional human feedback

Reference 32

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T19:02:26.113806Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 7e46c0fa-2736-4eaa-a853-aa8f29f2735d · outbound

This paper cites mplug-owl2: Revolutionizing multi-modal large language model with modality collaboration.

Beyond Global Editing: Per-Instance Disentangled Subspaces for Training-Free Hallucination Mitigation in LVLMs mplug-owl2: Revolutionizing multi-modal large language model with modality collaboration

Reference 33

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T19:02:26.151647Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 480d75e1-e6eb-43e5-ac11-b0679672a62f · outbound

This paper cites Analyzing and mitigating object hallucina- tion in large vision-language models.

Beyond Global Editing: Per-Instance Disentangled Subspaces for Training-Free Hallucination Mitigation in LVLMs Analyzing and mitigating object hallucina- tion in large vision-language models

Reference 34

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T19:02:26.096741Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 644e7172-8186-494e-bea1-a4e58f156d3a · outbound

This paper cites MiniGPT-4: Enhancing Vision-Language Understanding with Advanced Large Language Models.

Beyond Global Editing: Per-Instance Disentangled Subspaces for Training-Free Hallucination Mitigation in LVLMs MiniGPT-4: Enhancing Vision-Language Understanding with Advanced Large Language Models

Reference 35

Resolution: unresolved; no resolver link, observed 2026-08-11T19:01:40.520656Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 931176c0-a628-49ec-83f9-030f000bc6dc · outbound

This paper cites Analyzing and Mitigating Object Hallucination in Large Vision-Language Models.

Beyond Global Editing: Per-Instance Disentangled Subspaces for Training-Free Hallucination Mitigation in LVLMs Analyzing and Mitigating Object Hallucination in Large Vision-Language Models

Reference 36

Resolution: unresolved; no resolver link, observed 2026-08-11T19:01:40.508642Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 74c97226-a88a-470a-b317-6839b4b0809e · outbound

This paper cites Vasparse: Towards efficient visual hallucination mitigation via visual-aware token sparsification.

Beyond Global Editing: Per-Instance Disentangled Subspaces for Training-Free Hallucination Mitigation in LVLMs Vasparse: Towards efficient visual hallucination mitigation via visual-aware token sparsification

Reference 39

Resolution: unresolved; no resolver link, observed 2026-08-11T19:01:40.526644Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 51435154-9683-432d-b371-aeae1b732b78 · outbound

This paper cites an unresolved cited work.

Beyond Global Editing: Per-Instance Disentangled Subspaces for Training-Free Hallucination Mitigation in LVLMs Unresolved cited work

Reference 2024

Resolution: unresolved; raw_fallback, observed 2026-08-11T19:02:26.517913Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation c730f4a9-78e7-4466-8a08-a92de88186d8 · outbound

This paper cites URLhttps://bmva-archive.org.uk/bmvc/2025/ assets/papers/Paper_1137/paper.pdf.

Beyond Global Editing: Per-Instance Disentangled Subspaces for Training-Free Hallucination Mitigation in LVLMs URLhttps://bmva-archive.org.uk/bmvc/2025/ assets/papers/Paper_1137/paper.pdf

Reference 2025

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T19:02:26.459291Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Pith citing papers

No inbound Pith citation observations are available.