Pith. sign in

REVIEW 7 cited by

Multimodal Needle in a Haystack: Benchmarking Long-Context Capability of Multimodal Large Language Models

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 2406.11230 v2 pith:SEPQ5USW submitted 2024-06-17 cs.LG cs.AIcs.CLcs.CV

classification cs.LGcs.AIcs.CLcs.CV
keywords long-contextmllmsmodelsmultimodalimageapi-basedbenchmarkcapabilities
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Multimodal Large Language Models (MLLMs) have shown significant promise in various applications, leading to broad interest from researchers and practitioners alike. However, a comprehensive evaluation of their long-context capabilities remains underexplored. To address these gaps, we introduce the MultiModal Needle-in-a-haystack (MMNeedle) benchmark, specifically designed to assess the long-context capabilities of MLLMs. Besides multi-image input, we employ image stitching to further increase the input context length, and develop a protocol to automatically generate labels for sub-image level retrieval. Essentially, MMNeedle evaluates MLLMs by stress-testing their capability to locate a target sub-image (needle) within a set of images (haystack) based on textual instructions and descriptions of image contents. This setup necessitates an advanced understanding of extensive visual contexts and effective information retrieval within long-context image inputs. With this benchmark, we evaluate state-of-the-art MLLMs, encompassing both API-based and open-source models. The findings reveal that GPT-4o consistently surpasses other models in long-context scenarios, but suffers from hallucination problems in negative samples, i.e., when needles are not in the haystacks. Our comprehensive long-context evaluation of MLLMs also sheds lights on the considerable performance gap between API-based and open-source models. All the code, data, and instructions required to reproduce the main results are available at https://github.com/Wang-ML-Lab/multimodal-needle-in-a-haystack.

Discussion (0). Sign in to comment.

Forward citations

Cited by 7 Pith papers

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

  1. TS-Haystack: A Multi-Task Retrieval Benchmark for Long-Context Time-Series Reasoning

    cs.LG 2026-02 unverdicted novelty 7.0 of 10

    TS-Haystack benchmark shows time-series language models degrade sharply on long contexts while an agentic retrieval system using classifier tools matches or beats them on 9 of 10 tasks.

  2. Movie Facts and Fibs (MF$^2$): A Benchmark for Long Movie Understanding

    cs.CV 2025-06 conditional novelty 7.0 of 10

    MF2 evaluates long-movie understanding by asking models to classify fact/fib claim pairs; the best model trails humans by 23.5 points in pairwise accuracy.

  3. Where Facts Go Missing: A Layerwise Taxonomy and Per-Layer Attribution of Information Omission in Air-Gapped LLMAgent Pipelines

    cs.MA 2026-07 conditional novelty 6.0 of 10

    In a controlled 75,476-trial stress test, about 73% of omitted-fact failures in LLM agent pipelines are traced to deterministic middleware (redaction, pagination, truncation) rather than model behavior.

  4. HomeBench: Evaluating LLMs in Smart Homes with Valid and Invalid Instructions Across Single and Multiple Devices

    cs.CL 2025-05 conditional novelty 6.0 of 10

    HomeBench is a new smart home benchmark that exposes near-zero success rates for top LLMs on invalid multi-device instructions.

  5. CLBench-V: Evaluating Multimodal Context Learning from Grounding to Knowledge Acquisition

    cs.CV 2026-07 conditional novelty 5.0 of 10

    Multimodal context-learning benchmark CLBench-V separates grounding, information application, and knowledge acquisition; the best evaluated model scores 0.2847.

  6. Toward Embodied AGI: A Review of Embodied AI and the Road Ahead

    cs.AI 2025-05 accept novelty 4.0 of 10

    Embodied AI today sits between Level 1 and Level 2 on a new five-level roadmap toward all-purpose humanlike robots.

  7. Differential Multimodal Transformers

    cs.AI 2025-07 reject novelty 2.0 of 10

    A proposed multimodal extension of differential attention turns out to be equivalent to a learned scalar multiplier on standard attention, so the reported retrieval gains do not support the claimed mechanism.

Pith tools