HALDETECT at ImageEval 2026 Shared Tasks: Answer-First Contrastive Grounding with QLoRA

• URL: http://arxiv.org/abs/2609.11236v1
• Date: Thu, 10 Sep 2026 08:35:29 GMT
• Title: HALDETECT at ImageEval 2026 Shared Tasks: Answer-First Contrastive Grounding with QLoRA
• Abstract summary: We present HALDETECT, a system for the English hallucination-detection track (Task 1b) of ImageEval 2026.<n>We frame the item as one contrastive decision, emit the answer before its explanation, and structure reasoning around colour/texture, shape/form, and context.<n>Our best submitted adapter fine-tunes Qwen2.5-VL-7B-Instruct with 4-bit QLoRA while freezing the vision encoder and reaches Contrastive Instability (CI) 0.035 on the 1,000-item test set.
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Large multimodal models tend to hallucinate visual detail fluently, which limits their deployment for fine-grained interpretation. We present HALDETECT, our system for the English hallucination-detection track (Task 1b) of ImageEval 2026, in which a system must identify, from an image and three culturally plausible statements, the single visually grounded one. We frame the item as one contrastive decision, emit the answer before its explanation, and structure reasoning around colour/texture, shape/form, and context. Our best submitted adapter fine-tunes Qwen2.5-VL-7B-Instruct with 4-bit QLoRA while freezing the vision encoder and reaches Contrastive Instability (CI) 0.035 on the 1,000-item test set; we placed third of eight teams. Development experiments show that answer order can matter more than model scale and that adaptation beats prompting alone. Retrospective paired analysis of the released gold labels confirms the QLoRA gain over the best prompt but not the small gap between the devtest-selected and best-test adapters, and reseeding all four training sizes shows that the apparent data-scaling curve does not survive a seed change. The 35 residual errors are culturally plausible function, material, and recognition distinctions; naive adapter voting does not help.

Related papers

• SciTrue: Reliable Scientific Claim Validation with Frontier and Open Language Models at the NTCIR SciClaimEval Task [15.259880314614117]
  We benchmark eleven frontier and open multimodal models under one honest, per-sample protocol.<n>SciTrue placed first by a clear margin in three of the four evidence-category/subtask combinations.
  arXiv  Detail & Related papers  (2026-09-01T03:30:45Z)
• Answer-Conditioned Chain-of-Thought Distillation for Few-Shot Industrial Vision with Small VLMs [0.0]
  We propose answer-conditioned chain-of-thought (CoT) distillation for rapidly adapting small vision-language models to new industrial tasks.<n>A frontier VLM receives each training image along with its correct label and generates a justified visual explanation.<n>We validate on four industrial classification tasks spanning three image modalities using only 18 to 30 labeled images per task.
  arXiv  Detail & Related papers  (2026-07-12T09:23:05Z)
• VisualFLIP: Do Predictions Depend on Task-Critical Visual Evidence in Multimodal Reasoning? [79.44137632338062]
  When a multimodal large language model answers a visual reasoning question correctly, is the prediction actually supported by the task-critical visual evidence?<n>We introduce VisualFLIP, a paired benchmark with 1,374 images arranged as same-question pairs across cardinality, attribute, spatial, and logic tasks.<n>We evaluate 24 MLLMs with pair accuracy, which requires solving both sides of a pair, and Collapse Rate (CR), which measures how often a model that solves at least one side repeats the same non-empty answer for both images.
  arXiv  Detail & Related papers  (2026-06-05T22:06:07Z)
• NeuroQA: A Large-Scale Image-Grounded Benchmark for 3D Brain MRI Understanding [6.217658756255346]
  NeuroQA is a large-scale benchmark for visual question answering in 3D brain magnetic resonance imaging (MRI)<n>It spans ages 5-104 and five clinical domains: Alzheimer's, Parkinson's, tumors, white matter disease, and neurodevelopment.<n>NeuroQA evaluates 11 clinically grounded reasoning skills across Yes/No, multiple-choice, and open-ended formats.
  arXiv  Detail & Related papers  (2026-05-19T21:54:12Z)
• HopChain: Multi-Hop Data Synthesis for Generalizable Vision-Language Reasoning [86.82637240330791]
  Long chain-of-thought (CoT) reasoning exposes diverse failure modes.<n>Most existing vision-language data used for reinforcement learning with verifiable rewards does not involve complex reasoning chains.<n>We propose HopChain, a scalable framework for synthesizing multi-hop vision-language reasoning data.
  arXiv  Detail & Related papers  (2026-03-17T18:04:58Z)
• Contra4: Evaluating Contrastive Cross-Modal Reasoning in Audio, Video, Image, and 3D [97.08549913899247]
  We introduce Contra4, a dataset for contrastive cross-modal reasoning across four modalities: image, audio, video, and 3D.<n>Each example presents a natural language question alongside multiple candidate modality instances, and the model must select the one that semantically aligns with the prompt.<n>Contra4 combines human-annotated captions with a mixture-of-models round-trip-consistency filter to ensure high-quality supervision, resulting in 174k training examples and a manually verified test set of 2.3k samples.
  arXiv  Detail & Related papers  (2025-06-02T03:12:13Z)
• LENS: Multi-level Evaluation of Multimodal Reasoning with Large Language Models [59.0256377330646]
  Lens is a benchmark with 3.4K contemporary images and 60K+ human-authored questions covering eight tasks and 12 daily scenarios.<n>This dataset intrinsically supports to evaluate MLLMs to handle image-invariable prompts, from basic perception to compositional reasoning.<n>We evaluate 15+ frontier MLLMs such as Qwen2.5-VL-72B, InternVL3-78B, GPT-4o and two reasoning models QVQ-72B-preview and Kimi-VL.
  arXiv  Detail & Related papers  (2025-05-21T15:06:59Z)
• Multi-Modal Hallucination Control by Visual Information Grounding [121.6983694815504]
  We show that Generative Vision-Language Models (VLMs) are prone to generate plausible-sounding textual answers that are not always grounded in the input image. We introduce Multi-Modal Mutual-Information Decoding (M3ID), a new sampling method for prompt amplification. M3ID amplifies the influence of the reference image over the language prior, hence favoring the generation of tokens with higher mutual information with the visual prompt.
  arXiv  Detail & Related papers  (2024-03-20T22:05:18Z)
• HallusionBench: An Advanced Diagnostic Suite for Entangled Language Hallucination and Visual Illusion in Large Vision-Language Models [69.52245481329899]
  We introduce HallusionBench, a benchmark for the evaluation of image-context reasoning. The benchmark comprises 346 images paired with 1129 questions, all meticulously crafted by human experts. In our evaluation on HallusionBench, we benchmarked 15 different models, highlighting a 31.42% question-pair accuracy achieved by the state-of-the-art GPT-4V.
  arXiv  Detail & Related papers  (2023-10-23T04:49:09Z)
• Detecting and Preventing Hallucinations in Large Vision Language Models [4.7264116948935975]
  M-HalDetect is the first multi-modal hallucination detection dataset for detailed image descriptions. We train fine-grained multi-modal reward models from InstructBLIP and evaluate their effectiveness with best-of-n rejection sampling. We find that our reward model generalizes to other multi-modal models, reducing hallucinations in LLaVA and mPLUG-OWL by 15% and 57% respectively.
  arXiv  Detail & Related papers  (2023-08-11T21:35:20Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.