A new large-scale triplet dataset and diffusion transformer model using coarse human masks deliver improved video virtual try-on quality and generalization in challenging real-world conditions.
hub
Jena lectures on generalized global symmetries: Principles and applications, July 2024
12 Pith papers cite this work. Polarity classification is still indexing.
12
Pith papers citing it
hub tools
citation-role summary
background 4
citation-polarity summary
roles
background 4polarities
background 4representative citing papers
KernelSight-LM simulates LLM inference at kernel granularity with cross-generation (12.1% per-kernel error) and target-measured (3.8% error) tiers, yielding end-to-end median errors of 15.4%/12.8%/3.0% and 14.3%/6.2%/2.7% for TTFT/TPOT/throughput across six model families.
A multi-reference audit framework for LLM translations of the Pali Canon uses embedding drift from a human reference centroid to triage candidates for LLM-judge adjudication, showing drift correlates with major error rates and model-specific differences in the high-drift tail.
First unified survey formalizing Pretraining Data Exposure across exposure levels and reviewing attack, defense, and contamination methods for LLMs.
Refines charge quantization via homotopy type A whose homotopy groups classify brane charges and homology groups classify higher-form symmetries, deriving swampland-like constraints that rule out noncompact gauge groups and non-nilpotent Lie algebras for field strengths.
FASTER models multi-candidate denoising as an MDP and trains a value function to filter actions early, delivering the performance of full sampling at lower cost in diffusion RL policies.
LABBench2 is a more challenging benchmark than LAB-Bench for assessing AI performance on biology research tasks, with frontier models showing accuracy drops of 26-46% across subtasks.
SKA FRB dispersion measures and their cross-correlations with Stage IV shear and clustering can pin down baryonic feedback and improve cosmological constraints by factors of ~2–5 under optimistic detection rates.
GeneCS compiler reduces ancillary qubits and checks by over 85% on average for single- and cross-code logical operations on stabilizer codes while preserving error rates and scaling to over 10,000 qubits.
Eywa enables language-based agentic AI systems to collaborate with specialized scientific foundation models for improved performance on structured data tasks.
CG-MLLM is a multimodal LLM using a Mixture-of-Transformer architecture with separate TokenAR and BlockAR components integrated with a pre-trained vision-language backbone and 3D VAE to enable 3D captioning and high-fidelity generation.
citing papers explorer
-
TripVVT: A Large-Scale Triplet Dataset and a Coarse-Mask Baseline for In-the-Wild Video Virtual Try-On
A new large-scale triplet dataset and diffusion transformer model using coarse human masks deliver improved video virtual try-on quality and generalization in challenging real-world conditions.
-
KernelSight-LM: A Kernel-Level LLM Inference Simulator
KernelSight-LM simulates LLM inference at kernel granularity with cross-generation (12.1% per-kernel error) and target-measured (3.8% error) tiers, yielding end-to-end median errors of 15.4%/12.8%/3.0% and 14.3%/6.2%/2.7% for TTFT/TPOT/throughput across six model families.
-
From Outliers to Errors: Auditing Pali-to-English LLM Translations with Multi-Reference Adjudication
A multi-reference audit framework for LLM translations of the Pali Canon uses embedding drift from a human reference centroid to triage candidates for LLM-judge adjudication, showing drift correlates with major error rates and model-specific differences in the high-drift tail.
-
Pretraining Data Exposure in Large Language Models: A Survey of Membership Inference, Data Contamination, and Security Implications
First unified survey formalizing Pretraining Data Exposure across exposure levels and reviewing attack, defense, and contamination methods for LLMs.
-
Generalised Symmetries and Swampland-Type Constraints from Charge Quantisation via Rational Homotopy Theory
Refines charge quantization via homotopy type A whose homotopy groups classify brane charges and homology groups classify higher-form symmetries, deriving swampland-like constraints that rule out noncompact gauge groups and non-nilpotent Lie algebras for field strengths.
-
FASTER: Value-Guided Sampling for Fast RL
FASTER models multi-candidate denoising as an MDP and trains a value function to filter actions early, delivering the performance of full sampling at lower cost in diffusion RL policies.
-
LABBench2: An Improved Benchmark for AI Systems Performing Biology Research
LABBench2 is a more challenging benchmark than LAB-Bench for assessing AI performance on biology research tasks, with frontier models showing accuracy drops of 26-46% across subtasks.
-
Probing the Baryon Distribution with Fast Radio Bursts
SKA FRB dispersion measures and their cross-correlations with Stage IV shear and clustering can pin down baryonic feedback and improve cosmological constraints by factors of ~2–5 under optimistic detection rates.
-
GeneCS: Synthesizing Resource-Efficient Code Surgery for Arbitrary Quantum Stabilizer Codes
GeneCS compiler reduces ancillary qubits and checks by over 85% on average for single- and cross-code logical operations on stabilizer codes while preserving error rates and scaling to over 10,000 qubits.
-
Heterogeneous Scientific Foundation Model Collaboration
Eywa enables language-based agentic AI systems to collaborate with specialized scientific foundation models for improved performance on structured data tasks.
-
CG-MLLM: Captioning and Generating 3D content via Multi-modal Large Language Models
CG-MLLM is a multimodal LLM using a Mixture-of-Transformer architecture with separate TokenAR and BlockAR components integrated with a pre-trained vision-language backbone and 3D VAE to enable 3D captioning and high-fidelity generation.
- EyeMulator: Improving Code Language Models by Mimicking Human Visual Attention