Technical University of Munich · 14 records

dateidtitleauthorsvenue↑
2026-11-122611.12869Where Does Alignment Emerge and What Does It Mean? A Closer Look at the Platonic Representation HypothesisJulia Herbst et al.arXiv6
2026-11-042610.40574Every Token Needs Its Own Teacher: Multi-teacher Adaptive On-policy Self-distillationYifu Luo et al.arXiv13
2026-11-032610.32878Advancing SVD-based LLM Compression via Layer-Wise Error Model SearchMoritz Thoma et al.arXiv4
2026-10-222610.26397Bridged Self-Distillation for On-Policy LearningRunsheng Chen et al.arXiv4
2026-10-212610.27190On the Fragility of Tasks under Composition: Implicit Budgeting in Language ModelsChenpan Li et al.arXiv11
2026-09-282608.09444Depth-adaptive Inference of Looped Language Models via Continuous Depth BatchingKristian Schwethelm et al.arXiv9
2026-09-012608.29847ContextBias: Controlled Evaluation of Bias Persistence Under Context Shift in Text-to-Image ModelsShaghayegh Kolli et al.arXiv5
2026-08-272608.15062Gated Recurrent Transformers: Expressive Depth through Recurrent ModulationAmr Hegazy et al.arXiv11
2026-06-222606.20527StylisticBias: A Few Human Visual Cues Drive Most Social Biases in MLLMsShaghayegh Kolli et al.arXiv4
2026-06-042605.31559Functional Attention: From Pairwise Affinities to Functional CorrespondencesJiefang Xiao et al.arXiv2
2026-06-032606.03287BA-T: An Iterative Transformer for Two-View Bundle AdjustmentGanlin Zhang et al.arXiv1
2026-05-142605.13542RealICU: Do LLM Agents Understand Long-Context ICU Data? A Benchmark Beyond Behavior ImitationChengzhi Shen et al.arXiv6
2026-05-142605.10315Active Tabular Augmentation via Policy-Guided Diffusion InpaintingZheyu Zhang et al.arXiv
2026-05-042605.00414Trees to Flows and Back: Unifying Decision Trees and Diffusion ModelsSai Niranjan Ramachandran et al.arXiv7