11 papers · 1 filter
Physics as the label for measuring and correcting materials reasoning in multimodal models
Hasan Kurban, Rasul Khanbayov, Mustafa Kurban
Vision-language and language models increasingly interpret materials data, yet benchmarks report that they hallucinate invalid properties and violate physical law. Evaluation match…
A Low Grounding Score Is Not an Ungrounded Judge: Identifying the Perceptibility Confound in Multimodal Oversight
Rasul Khanbayov, Hasan Kurban
Model judges now supervise multimodal systems at scale, filtering training data, selecting outputs, and supplying the reward that shapes multimodal reasoning models. Trusting one m…
PhysWeep: Does a Video Generator Realize the Physics You Ask For?
Rasul Khanbayov, Hasan Kurban
Image-to-video generators are often credited with absorbing physical dynamics as implicit world models, a claim the community currently checks with plausibility scores that ask whe…
One Perturbation Is Not Enough: Identifiability and Blind Baselines for Behavioral AI Evaluation
Rasul Khanbayov, Mariam Sohail, Ahmed Abdala +1
Behavioral evaluations perturb an input and read the induced change in the output in order to certify that a system uses that input. We show that the number of perturbations such a…
Accuracy and Cost Claims Do Not Survive Re-Execution in Agentic VideoQA
Rama AlHamidi, Aseel Mohamed, Rasul Khanbayov +3
Agentic Video Question Answering (VideoQA) systems produce answers through adaptive reasoning and tool-use trajectories, yet standard practice evaluates each system once and estima…
Counterfactual Sensitivity Is Not Repairability: Auditing Replay Probes for Video Evidence
Rama AlHamidi, Rasul Khanbayov, Erchin Serpedin +1
Tool-using video agents retrieve visual evidence before answering, but the final answer is not forced to depend on what was retrieved. The natural black box test is counterfactual:…
Conformal Coverage Guarantees for Any Video Temporal Grounder
Aseel Mohamed, Rasul Khanbayov, Erchin Serpedin +1
Event boundaries in continuous video are ambiguous: re-annotate the same query-video pair and independent annotators mark moments that overlap by less than half on a large fraction…
GRAPE: Graph-Augmented Prototype Explanations for Interactive Medical Image Diagnosis
Rasul Khanbayov, Erchin Serpedin, Hasan Kurban
Prototype-based medical image classifiers present three clinical limitations: they treat findings as independent, silently amplify unsafe physician feedback, and require full retra…
4D Synchronized Fields: Motion-Language Gaussian Splatting for Temporal Scene Understanding
Mohamed Rayan Barhdadi, Samir Abdaljalil, Rasul Khanbayov +2
Current 4D representations decouple geometry, motion, and semantics: reconstruction methods discard interpretable motion structure; language-grounded methods attach semantics after…
IRIS: A Real-World Benchmark for Inverse Recovery and Identification of Physical Dynamic Systems from Monocular Video
Rasul Khanbayov, Mohamed Rayan Barhdadi, Erchin Serpedin +1
Unsupervised physical parameter estimation from video lacks a common benchmark: existing methods evaluate on non-overlapping synthetic data, the sole real-world dataset is restrict…
Automated Knot Detection and Pairing for Wood Analysis in the Timber Industry
Guohao Lin, Shidong Pan, Rasul Khanbayov +3
Knots in wood are critical to both aesthetics and structural integrity, making their detection and pairing essential in timber processing. However, traditional manual annotation wa…