💾 CPU Caching - emschwartz · Scour

Dual Alignment Between Language Model Layers and Human Sentence Processing 🔤Tokenization

Cat-DPO: Category-Adaptive Safety Alignment 🛡️AI Safety

S2H-DPO: Hardness-Aware Preference Optimization for Vision-Language Models 📊Embeddings

Bridging the Gap between User Intent and LLM: A Requirement Alignment Approach for Code Generation 🪄Prompt Engineering

SaFeR-Steer: Evolving Multi-Turn MLLMs via Synthetic Bootstrapping and Feedback Dynamics 🛡️AI Safety

Into the Gray Zone: Domain Contexts Can Blur LLM Safety Boundaries 🛡️AI Safety

Adaptive Forensic Feature Refinement via Intrinsic Importance Perception 🔍FAISS

The Rise of Verbal Tics in Large Language Models: A Systematic Analysis Across Frontier Models 🛡️AI Safety

Learning to Credit the Right Steps: Objective-aware Process Optimization for Visual Generation ✨Gemini

SignDPO: Multi-level Direct Preference Optimisation for Skeleton-based Gloss-free Sign Language Translation 🔤Tokenization

Jailbreaking Large Language Models with Morality Attacks 🛡️AI Security

Discovering a Shared Logical Subspace: Steering LLM Logical Reasoning via Alignment of Natural-Language and Symbolic Views 🧠LLM Inference

Involuntary In-Context Learning: Exploiting Few-Shot Pattern Completion to Bypass Safety Alignment in GPT-5.4 🛡️AI Safety

AIFIND: Artifact-Aware Interpreting Fine-Grained Alignment for Incremental Face Forgery Detection 🔥Burn

The Triadic Loop: A Framework for Negotiating Alignment in AI Co-hosted Livestreaming 🌐ActivityPub Protocol

CS3: Efficient Online Capability Synergy for Two-Tower Recommendation 🎲Recommendation Systems

Log in to enable infinite scrolling