large language modelslegal liability adjudicationpublished Nov 6, 2026 · arXiv:2611.06972
KDASO: Knowledge-Data Dual-Driven Automated Skill Optimization for Liability Adjudication Tasks
federated fine-tuninglarge language modelspublished Nov 6, 2026 · arXiv:2611.07598
Align Before Aggregation: Basis-Consistent Federated LoRA under Heterogeneous Ranks
decentralized LLM teamstask allocationpublished Nov 2, 2026 · arXiv:2610.41044
Consensus Within, Competition Between: Decentralized LLM Teams for Perimeter Defense
toxicity reductionrepresentation steeringpublished Nov 2, 2026 · arXiv:2610.41165
LLM Detoxification via Transition-Aware Reward-Guided Representation Sampling
belief updatesagent safetypublished Oct 31, 2026 · arXiv:2610.38869
Auditable Belief Updates as a Missing Safety Layer for LLM-Powered Multi-Agent Systems
causal reasoningbenchmark evaluationpublished Oct 31, 2026 · arXiv:2610.38100
Evaluating Causal Reasoning in LLMs: A Case Study in the Challenges of Procedural Generation