LoopGuard: Breaking Self-Reinforcing Attention Loops via Dynamic KV Cache Interv

arXiv:2604.05887v1 Announce Type: new Abstract: Multimodal Large Language Models (MLLMs) have advanced unified reasoning over text, images, and videos, but their inference is

arXiv:2610.02815v1 Announce Type: new Abstract: Long chain-of-thought reasoning substantially increases KV-cache memory during autoregressive decoding, as every generated token

arXiv:2608.20617v1 Announce Type: new Abstract: Multi-agent LLM systems split work across models, so answering often requires knowledge that sits in another agent's context: a

arXiv:2606.13705v1 Announce Type: cross Abstract: Yes. Can it cure doom loops? Probably not. The Gemma 4 instruction-tuned models share a reproducible failure: on long factual

arXiv:2604.05013v1 Announce Type: cross Abstract: Current LLM coding agents are predominantly trained on composite benchmarks (e.g., bug fixing), which often leads to

arXiv:2603.22300v1 Announce Type: cross Abstract: Scaling Transformers to ultra-long contexts is bottlenecked by the $O(n^2 d)$ cost of self-attention. Existing methods reduce

arXiv:2608.13043v1 Announce Type: new Abstract: Diffusion models have achieved dominant performance in visual generation but suffer from substantial inference overhead. While