Large Language Models

Route to Rome Attack: Directing LLM Routers to Expensive Models via Adversarial Suffix Optimization ACL 2026 Routing with Generated Data: Annotation-Free LLM Skill Estimation and Expert Selection ACL 2026 RRAtention: Dynamic Block Sparse Attention via Per-Head Round-Robin Shifts for Long-Context Inference ACL 2026 RSDA: Restoring Stale Data Affinity via Dynamic Renovation Strategy for Mitigating Data Scarcity ACL 2026 rSIM: Incentivizing Reasoning Capabilities of LLMs via Reinforced Strategy Injection ACL 2026 RubricBench: Aligning Model-Generated Rubrics with Human Standards ACL 2026 S2O: Early Stopping for Sparse Attention via Online Permutation ACL 2026 SACTOR: LLM-Driven Correct and Idiomatic C to Rust Translation with Static Analysis and FFI-Based Verification ACL 2026 SADA: Bridging In-Context Learning and Fine-Tuning via State-Aligned Distillation Adapters ACL 2026 SafeAgent: Safeguarding LLM Agents via an Automated Risk Simulator ACL 2026 SafeConstellations: Mitigating Over-Refusals in LLMs Through Task-Aware Representation Steering ACL 2026 Safe-FedLLM: Delving into the Safety of Federated Large Language Models ACL 2026 SafeGRPO: Self-Rewarded Multimodal Safety Alignment via Rule-Governed Policy Optimization CVPR 2026 Safeguarding LLM Fine-tuning via Push-Pull Distributional Alignment ACL 2026 Safeguarding LLM Training at Scale: Online SDC Detection and Insights from 35 Million GPU Hours OSDI 2026 SafeMT: Multi-turn Safety for Multimodal Language Models ACL 2026 SafetyMem: Adaptive Jailbreak Defense via Dual-Component Safety Memory ACL 2026 Safety-Utility Conflicts Are Not Global: Surgical Alignment via Head-Level Diagnosis ACL 2026 SAFO: Stable Adaptive Fairness Optimization for LLM-Based Social Survey Simulation ACL 2026 SAGE: Sparse Adaptive Guidance for Dependency-Aware Tabular Data Generation ACL 2026 SAIR-Comb : A Structure-Aware Iterative Refinement Framework for Combinatorics Autoformalization ACL 2026 SAJA: A Simple Approach to Judge Alignment for LLM-as-a-Judge ACL 2026 Same Benchmark, Same Subspace: Task-Selective Convergence in LLM Representations UAI 2026 SARA: Selective and Adaptive Retrieval-augmented Generation with Context Compression ACL 2026 SARL-STG: A Spatially Aware Reinforcement Learning Framework for Refining MLLMs in Spatio-Temporal Video Grounding CVPR 2026