SDGBiasBench: Benchmarking and Mitigating Vision--Language Models' Biases in Sustainable Development Goals

• URL: http://arxiv.org/abs/2605.21919v1
• Date: Thu, 21 May 2026 02:44:54 GMT
• Title: SDGBiasBench: Benchmarking and Mitigating Vision--Language Models' Biases in Sustainable Development Goals
• Abstract summary: Assessment of progress toward the Sustainable Development Goals requires multi-step reasoning over visual cues, contextual knowledge, and development indicators.<n>Existing benchmarks typically evaluate these aspects in isolation, obscuring systematic biases that emerge when models substitute priors for evidence.<n>We propose SDGBiasBench, a large-scale benchmark suite for vision-language reasoning.
• Score: 28.710989810208748
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Assessing progress toward the Sustainable Development Goals (SDGs) requires multi-step reasoning over visual cues, contextual knowledge, and development indicators, where incomplete evidence use and imperfect evidence integration can introduce hidden prediction biases. Real-world SDG monitoring further spans both qualitative judgments and quantitative estimation. However, existing benchmarks typically evaluate these aspects in isolation, obscuring systematic biases that emerge when models substitute priors for evidence. To address this gap, we propose SDGBiasBench, a large-scale benchmark suite for SDG-oriented vision-language reasoning. Spanning 500k expert-involved multiple-choice questions and 50k regression tasks, the benchmark enables comprehensive assessment of both decision-level and estimation-level bias in Vision--Language Models (VLMs). Evaluations on SDGBiasBench reveal an intrinsic SDG bias in current VLMs, where predictions are frequently driven by SDG specific priors rather than reliable multi-modal cues. To mitigate such bias, we propose CADE (Contrastive Adaptive Debias Ensemble), a training-free, plug-and-play method that leverages modality-specific answer priors. CADE yields significant gains on the proposed benchmark, improving multiple-choice accuracy by up to 25% and reducing regression MAE by up to 12 points across multiple VLMs. We hope our work can foster the development of more fair and reliable AI systems for sustainable development.

Related papers

• COSMO: Consensus-Driven Shift Modulation for Source-Free Domain Adaptation [79.43842208099619]
  Source-free domain adaptation (SFDA) adapts a source-trained model to an unlabeled target domain without source data.<n>We formulate VLM-guided SFDA as a sample-wise reliability-allocation problem.<n>We propose Consensus-Driven Shift Modulation (COSMO) to replace expert-to-expert guidance with co-adaptation.
  arXiv  Detail & Related papers  (2026-08-05T09:10:30Z)
• SD-MAR: Multi-image Analytical Reasoning via Synthetic Data and Reinforcement Learning [13.79533690995798]
  We introduce SD-MAR, a framework for training and evaluating Vision Language Models (VLMs) on multi-image analytical reasoning.<n> SD-MAR constructs paired visual scenarios through controlled perturbations and generates reasoning tasks spanning semantic change attribution and quantitative comparison.<n>We show that GRPO-lite fine-tuning on SD-MAR improves in-domain accuracy by up to 36.95%, with Qwen2.5-VL-7B outperforming GPT-4.1 on the SD-MAR benchmark.
  arXiv  Detail & Related papers  (2026-07-15T19:55:01Z)
• DeepBias: Adaptive In-depth Probing of Social Biases in LVLMs [71.85279836937578]
  We introduce DeepBias, an adaptive framework for the in-depth probing of social biases in LVLMs with carefully designed agents.<n>We build a benchmark named DeepBiasBench using our framework. By employing an ensemble of five diverse state-of-the-art LVLMs as anchors, the benchmark captures vulnerabilities shared across architectures.
  arXiv  Detail & Related papers  (2026-07-13T08:19:09Z)
• URAG: A Benchmark for Uncertainty Quantification in Retrieval-Augmented Large Language Models [35.441039437111606]
  URAG is a benchmark designed to assess the uncertainty of RAG systems across various fields like healthcare, programming, science, math, and general text.<n>We apply the evaluation pipeline to 8 standard RAG methods, measuring their performance through both accuracy and prediction-set sizes based on LAC and APS metrics.
  arXiv  Detail & Related papers  (2026-03-02T00:22:06Z)
• Beyond Majority Voting: Towards Fine-grained and More Reliable Reward Signal for Test-Time Reinforcement Learning [12.354777054071379]
  Test-time reinforcement learning mitigates reliance on annotated data by using majority voting results as pseudo-labels.<n>This voting strategy often induces confirmation bias and suffers from sparse rewards, limiting the overall performance.<n>We propose subgroup-specific step-wise confidence-weighted pseudo-label estimation (SCOPE) to address these issues.
  arXiv  Detail & Related papers  (2025-12-17T07:21:54Z)
• Alleviating Choice Supportive Bias in LLM with Reasoning Dependency Generation [8.918979781532036]
  We present Reasoning Dependency Generation (RDG), a novel framework for generating unbiased reasoning data.<n>RDG automatically constructs balanced reasoning QA pairs, explicitly (un)modeling the dependencies between choices, evidences, and justifications.<n>Experiments show that LLMs fine-tuned on RDG-generated data demonstrate a 81.5% improvement in memory-based experiments and 94.3% improvement in the evaluation-based experiment.
  arXiv  Detail & Related papers  (2025-11-28T08:52:05Z)
• Reliable and Reproducible Demographic Inference for Fairness in Face Analysis [63.46525489354455]
  We propose a fully reproducible DAI pipeline that replaces conventional end-to-end training with a modular transfer learning approach.<n>We audit this pipeline across three dimensions: accuracy, fairness, and a newly introduced notion of robustness, defined via intra-identity consistency.<n>Our results show that the proposed method outperforms strong baselines, particularly on ethnicity, which is the more challenging attribute.
  arXiv  Detail & Related papers  (2025-10-23T12:22:02Z)
• Certainty-Guided Reasoning in Large Language Models: A Dynamic Thinking Budget Approach [0.15749416770494704]
  We show that Certainty-Guided Reasoning (CGR) improves baseline accuracy while reducing token usage.<n>CGR can eliminate millions of tokens in aggregate, with tunable trade-offs between certainty thresholds and efficiency.<n>By integrating confidence into the reasoning process, CGR makes large reasoning language models more adaptive, trustworthy, and resource efficient.
  arXiv  Detail & Related papers  (2025-09-09T14:57:15Z)
• Revisiting Reliability in the Reasoning-based Pose Estimation Benchmark [27.134554623769898]
  The reasoning-based pose estimation (RPE) benchmark has emerged as a widely adopted evaluation standard for pose-aware large language models (MLLMs)<n>We identified critical and benchmark-quality issues that hinder fair and consistent quantitative evaluations.
  arXiv  Detail & Related papers  (2025-07-17T17:33:11Z)
• MMIE: Massive Multimodal Interleaved Comprehension Benchmark for Large Vision-Language Models [71.36392373876505]
  We introduce MMIE, a large-scale benchmark for evaluating interleaved multimodal comprehension and generation in Large Vision-Language Models (LVLMs)<n>MMIE comprises 20K meticulously curated multimodal queries, spanning 3 categories, 12 fields, and 102 subfields, including mathematics, coding, physics, literature, health, and arts.<n>It supports both interleaved inputs and outputs, offering a mix of multiple-choice and open-ended question formats to evaluate diverse competencies.
  arXiv  Detail & Related papers  (2024-10-14T04:15:00Z)
• Anticipating the Unseen Discrepancy for Vision and Language Navigation [63.399180481818405]
  Vision-Language Navigation requires the agent to follow natural language instructions to reach a specific target. The large discrepancy between seen and unseen environments makes it challenging for the agent to generalize well. We propose Unseen Discrepancy Anticipating Vision and Language Navigation (DAVIS) that learns to generalize to unseen environments via encouraging test-time visual consistency.
  arXiv  Detail & Related papers  (2022-09-10T19:04:40Z)
• General Greedy De-bias Learning [163.65789778416172]
  We propose a General Greedy De-bias learning framework (GGD), which greedily trains the biased models and the base model like gradient descent in functional space. GGD can learn a more robust base model under the settings of both task-specific biased models with prior knowledge and self-ensemble biased model without prior knowledge.
  arXiv  Detail & Related papers  (2021-12-20T14:47:32Z)

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.