Paper page - When No Benchmark Exists: Validating Comparative LLM Safety Scoring Without Ground-Truth Labels
… The substantial differences arise upstream of the chain, in claim-contract enforcement and deployment fit. …
Search
Paper page - PREPING: Building Agent Memory without Tasks
… AI-generated summary Agent memory is typically constructed either offline from curated demonstrations or online from post-deployment interactions. …
Paper page - FastKernels: Benchmarking GPU Kernel Generation in Production
… FastKernels doubles as a minimalistic, production-grade inference framework that runs at parity with hardened systems such as vLLM and SGLang on mainstream LLM serving and substantially exceeds upstream references on under-served architectures; each task's interface mirrors the corresponding module… …
Paper page - DECO: Sparse Mixture-of-Experts with Dense-Comparable Performance on End-Side Devices
… AI-generated summary While Mixture-of-Experts MoE scales model capacity without proportionally increasing computation, its massive total parameter footprint creates significant storage and memory-access bottlenecks, which hinder efficient end-side deployment that simultaneously requires high perfor… …
Paper page - BlenderRAG: High-Fidelity 3D Object Generation via Retrieval-Augmented Code Synthesis
… By retrieving semantically similar examples during generation, BlenderRAG improves compilation success rates from 40.8% to 70.0% and semantic normalized alignment from 0.41 to 0.77 CLIP similarity across four state-of-the-art LLMs, without requiring fine-tuning or specialized hardware, making it im… …
Paper page - Motion-Aware Caching for Efficient Autoregressive Video Generation
… AI-generated summary Autoregressive video generation paradigms offer theoretical promise for long video synthesis, yet their practical deployment is hindered by the computational burden of sequential iterative denoising . …
Paper page - Models That Know How Evaluations Are Designed Score Safer
… AI-generated summary The validity of AI safety evaluations depends on models behaving consistently across controlled and deployment settings. …
Paper page - FlashRT: Towards Computationally and Memory Efficient Red-Teaming for Prompt Injection and Knowledge Corruption
… However, security remains a major concern for their widespread deployment, with threats such as prompt injection and knowledge corruption . …
Paper page - MatryoshkaLoRA: Learning Accurate Hierarchical Low-Rank Representations for LLM Fine-Tuning
… AI-generated summary With the rise in scale for deep learning models to billions of parameters, the computational cost of fine-tuning remains a significant barrier to deployment. …
Paper page - MAIC-UI: Making Interactive Courseware with Generative UI
… A three-month classroom deployment with 53 high school students demonstrates that MAIC-UI fosters learning agency and reduces outcome disparities -- the pilot class achieved 9.21-point gains in STEM subjects compared to -2.32 points in control classes. …