arXiv Computer Vision
MotionBlind: Probing the Illusion of Motion Understanding in Video-LLMs
Read the original on arXiv Computer Vision →The Flow has not summarised this story yet — read it at arXiv Computer Vision.
Related stories
arXiv Computer Vision
Aug 28
R2M-Bench: Evaluating Revisit Memory via Relative Consistency in Interactive Video World Models
R2M-Bench is a benchmark that evaluates revisit memory in interactive video world models by comparing a revisit pair to two control pairs from the same rollout: a gap‑matched non‑revisit pair and a short‑range pair. It introduces MemoryGain (MG) and Normalized Memory Ratio (NMR) to quantify the revisit advantage over generic temporal stability and normalize it by short‑to‑baseline dynamics. Across 300 instances and seven models, NMR correlates with human judgments and reduces the influence of slow‑motion artifacts, with DreamX‑World‑Memo achieving the highest NMR.
arXiv AI
Jul 20
Think at 5 Hz, Act at 20 Hz: Asynchronous Fast-Slow Vision-Language-Action Inference for Closed-Loop Driving
arXiv:2607. 15621v1 Announce Type: cross Abstract: Large language models bring instruction following and scene reasoning to end-to-end driving, but their inference latency collides with the control rate a vehicle requires.
arXiv Computation and Language
Sep 4
Select, Compress, Reinvest: A Controlled Study of Visual-Token Allocation in Long-Video MLLMs
The study investigates how long‑video language models decide which frames to keep, compress, and reuse, testing each decision in isolation across six selection rules, three benchmarks, and two answering models. It finds that selecting frames based on queries yields the biggest performance boost, that halving spatial resolution costs little, and that reallocating saved tokens to more compressed frames can further improve accuracy. The work also highlights the importance of a unified evaluation harness to avoid misleading comparisons.
arXiv AI
Jul 16
Accuracy Without Grounding: Diagnosing Visual Dependency Dissociation in Video LLM Benchmarks
arXiv:2607. 13305v1 Announce Type: cross Abstract: Benchmark accuracy in video large language models (LLMs) is often treated as evidence of visual understanding.
arXiv AI
Sep 10
CALIPER: Clean Scenes Cannot Rank Physical Inference in Pretrained Visual Representations
CALIPER is a new benchmark that tests whether pretrained visual encoders can infer physical properties such as mass and friction from images. The test involves striking an object twice at known speeds, showing a third strike only up to contact, and asking a linear readout on frozen features to predict how far the object slides. Results show that in clean, fixed‑camera scenes all representations perform similarly, but when camera, lighting, and clutter are varied, only encoders that truly infer physics—like V‑JEPA 2—maintain performance, while random or raw pixel representations fail.
arXiv AI
Sep 10
TimeBlind: A Spatio-Temporal Compositionality Benchmark for Video LLMs
TimeBlind is a diagnostic benchmark designed to evaluate fine‑grained spatio‑temporal compositionality in video large language models (LLMs). It categorizes temporal understanding into three levels—atomic event recognition, event property characterization, and reasoning about event interdependencies—and uses a minimal‑pairs paradigm where video pairs share identical static content but differ only in temporal structure. Across 20 state‑of‑the‑art MLLMs tested on 600 curated instances, the best model achieved only 48.2% instance accuracy, far below human performance of 98.2%, highlighting a reliance on static visual shortcuts rather than true temporal reasoning.