Pith. sign in

REVIEW 19 cited by

VisualPuzzles: Decoupling Multimodal Reasoning Evaluation from Domain Knowledge

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2504.10342 v3 pith:SPM2RPYV submitted 2025-04-14 cs.CL

VisualPuzzles: Decoupling Multimodal Reasoning Evaluation from Domain Knowledge

classification cs.CL
keywords reasoningvisualpuzzlesknowledgebenchmarksmultimodalevaluatemodelsperformance
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
read the original abstract

Current multimodal benchmarks often conflate reasoning with domain-specific knowledge, making it difficult to isolate and evaluate general reasoning abilities in non-expert settings. To address this, we introduce VisualPuzzles, a benchmark that targets visual reasoning while deliberately minimizing reliance on specialized knowledge. VisualPuzzles consists of diverse questions spanning five categories: algorithmic, analogical, deductive, inductive, and spatial reasoning. One major source of our questions is manually translated logical reasoning questions from the Chinese Civil Service Examination. Experiments show that VisualPuzzles requires significantly less intensive domain-specific knowledge and more complex reasoning compared to benchmarks like MMMU, enabling us to better evaluate genuine multimodal reasoning. Evaluations show that state-of-the-art multimodal large language models consistently lag behind human performance on VisualPuzzles, and that strong performance on knowledge-intensive benchmarks does not necessarily translate to success on reasoning-focused, knowledge-light tasks. Additionally, reasoning enhancements such as scaling up inference compute (with "thinking" modes) yield inconsistent gains across models and task types, and we observe no clear correlation between model size and performance. We also found that models exhibit different reasoning and answering patterns on VisualPuzzles compared to benchmarks with heavier emphasis on knowledge. VisualPuzzles offers a clearer lens through which to evaluate reasoning capabilities beyond factual recall and domain knowledge.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 19 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Zone of Proximal Policy Optimization: Teacher in Prompts, Not Gradients

    cs.CL 2026-06 unverdicted novelty 7.0

    ZPPO improves distillation to small vision-language models by using binary and negative candidate prompts plus a replay buffer for hard questions, outperforming standard distillation and GRPO on a 31-benchmark suite w...

  2. StemBind: When MLLMs Get Lost Between Rules and Instances in Abstract Visual Reasoning

    cs.CV 2026-05 unverdicted novelty 7.0

    StemBind benchmark diagnoses MLLM failures in abstract visual reasoning by separating perception, rule induction, and answer selection on shared stems, finding a persistent rule-to-instance binding gap even when perce...

  3. TraversalBench: Challenging Paths to Follow for Vision Language Models

    cs.CV 2026-04 unverdicted novelty 7.0

    TraversalBench shows self-intersections cause the sharpest performance drops for VLMs on exact path traversal, with errors localized at the first crossing.

  4. Beacon: Knowing When and How to Perform Agentic Visual Reasoning

    cs.CV 2026-07 conditional novelty 6.0

    Beacon improves agentic visual reasoning by teaching models when tools are necessary and how to use them for net gains, via necessity-aware rewards and hint-guided RL.

  5. Trace: A Taxonomy-Guided Environment for Multidomain Visual Reasoning

    cs.CV 2026-07 conditional novelty 6.0

    RLVR training on 64,000 procedurally generated Trace instances improves Qwen2.5-VL macro-average on 24 external visual reasoning benchmarks by 3.51 points at 3B and 4.06 points at 7B.

  6. PACE: A Proxy for Agentic Capability Evaluation

    cs.AI 2026-07 conditional novelty 6.0

    A dual local/global selection of ~100 non-agentic instances predicts four agentic benchmarks with LOOCV MAE under 4%, Spearman above 0.80, and ~85% pairwise ranking accuracy at under 1% of full agent cost.

  7. PACE: A Proxy for Agentic Capability Evaluation

    cs.AI 2026-07 unverdicted novelty 6.0

    PACE builds proxy benchmarks from non-agentic instances via relevance and global selection plus regression to predict agentic scores with MAE under 4%, Spearman correlation above 0.80, and 85% ranking accuracy at unde...

  8. Self-Consistent Latent Reasoning: Long Latent Sequence Reasoning for Vision-Language Model

    cs.CV 2026-05 unverdicted novelty 6.0

    SCOLAR fixes information gain collapse in latent visual reasoning by generating independent auxiliary visual tokens via a detransformer, extending acceptable CoT length over 30x and delivering +14.12% gains on reasoni...

  9. Self-Consistent Latent Reasoning: Long Latent Sequence Reasoning for Vision-Language Model

    cs.CV 2026-05 unverdicted novelty 6.0

    SCOLAR addresses information gain collapse in latent visual reasoning by generating independent auxiliary visual tokens from LLM hidden states, extending acceptable CoT length over 30x and achieving +14.12% gains on b...

  10. Reinforcing Multimodal Reasoning Against Visual Degradation

    cs.CV 2026-05 unverdicted novelty 6.0

    ROMA improves MLLM robustness to seen and unseen visual corruptions by +2.3-2.4% over GRPO on seven reasoning benchmarks while matching clean accuracy.

  11. TraversalBench: Challenging Paths to Follow for Vision Language Models

    cs.CV 2026-04 conditional novelty 6.0

    VLMs fail at exact visual path following primarily at self-intersections, where performance drops sharply after the first crossing on a controlled polyline benchmark.

  12. MapTab: A Diagnostic Benchmark for Long-Horizon Multi-Criteria Multimodal Reasoning on Heterogeneous Topological Graphs

    cs.LG 2026-02 unverdicted novelty 6.0

    MapTab benchmark shows current MLLMs struggle with multi-criteria multimodal route planning and that combining vision and language frequently underperforms single-modality approaches.

  13. MapTab: A Diagnostic Benchmark for Long-Horizon Multi-Criteria Multimodal Reasoning on Heterogeneous Topological Graphs

    cs.LG 2026-02 conditional novelty 6.0

    MapTab introduces a 328-map, 196,800-query benchmark showing that current multimodal LLMs fall far short on multi-criteria route planning from maps-plus-tables.

  14. MapTab: A Diagnostic Benchmark for Long-Horizon Multi-Criteria Multimodal Reasoning on Heterogeneous Topological Graphs

    cs.LG 2026-02 conditional novelty 6.0

    MapTab is a new multimodal benchmark with 328 images and nearly 200k queries that shows current MLLMs have substantial difficulty with multi-criteria route planning when visual and tabular information must be combined.

  15. Look Light, Think Heavy: What Multimodal Chain-of-Thought Reasoning Can and Cannot Do

    cs.CL 2026-06 unverdicted novelty 5.0

    Multimodal CoT improves math and science reasoning but degrades visual perception performance and exhibits a pattern of diminishing visual reflection while verbal reflection varies.

  16. Semantic-Enriched Latent Visual Reasoning

    cs.CV 2026-05 unverdicted novelty 5.0

    SLVR is a two-stage method that enriches region-centric latent representations with fine-grained attribute semantics and aligns them via M-GRPO across multiple queries on the same region, supported by new SLV-Set data...

  17. Semantic-Enriched Latent Visual Reasoning

    cs.CV 2026-05 unverdicted novelty 5.0

    SLVR enriches latent visual representations with fine-grained attribute semantics via supervised first-stage learning and multi-query alignment via M-GRPO, yielding improved robustness on region-level reasoning tasks.

  18. SALLIE: Safeguarding Against Latent Language & Image Exploits

    cs.CR 2026-04 unverdicted novelty 5.0

    SALLIE detects jailbreaks in text and vision-language models by extracting residual stream activations, scoring maliciousness per layer with k-NN, and ensembling predictions, outperforming baselines on multiple datasets.

  19. Limits of Spatial Imagery Reasoning in Frontier LLM Models

    cs.CV 2026-03 unverdicted novelty 5.0

    An external 3D imagery tool does not fix frontier MLLM mental-rotation performance (≤62.5%), revealing missing visual-spatial primitives rather than a pure working-memory limit.