Penalty-Framed No-Valid-Option MCQA: Analyzing LLM Abstention under Invalid Choices
Research
Research News
New papers, breakthroughs and academic AI research
STRUCTURALCOST: A controlled reading time dataset for modeling human sentence processing difficulty
Research
Foveated Compression: Selective High-Resolution Preservation for Token-Efficient VLMs
Research
Image-Space Refraction Correction for Underwater 3D Reconstruction: Warping Flat-Port Views into Pinhole Perspective
Research
Towards benchmarking Western Bluebird detection in the wild
Research
Making COMET Comparable Across Scripts: Diagnosis and Correction of Tokeniser-Induced Script Bias in Indic MT Evaluation
Research
Language Unalignability: Why Some Concepts Resist Cross-Cultural Benchmark Evaluation
Research
AnyGroundBench: A Multi-Domain Adaptation Benchmark for Video Grounding in VLMs
Research
UI2App: Benchmarking Visual Interaction Inference in Executable Web Application Generation
Research
Memory Depth and Reconstructed Context Width: A Controlled Evaluation of Hierarchical Retrieval
Research
Knowing When Not to Answer: Cross-Domain and Multi-Turn Generalization of Latent Underspecification Signals
Research
Efficient Gaussian Splatting Sequence Compression with Standard Video Codecs
Research