Papers

AI papers from arXiv and journals, each with a summary and why it matters · 7705 papers
Oct 27
PapersBuildersMulti-sourceSelected10:42
Cachelint Detects Silent Prompt-Caching Bugs in LLM Agent Systems

The non-obvious takeaway is that caching bugs produce no errors at all — teams simply pay more per token while assuming caching works. Platform and backend engineers running agent workloads on OpenAI, Anthropic or other provider APIs can run CACHELINT on their own repos before costs accumulate. As an academic tool study with no pricing or regional availability angle, its closest comparison is other static-analysis approaches like CodeQL-based linting, which CACHELINT extends with provider-aware precondition checks.

Event
arXiv: Anthropic@Ying Zhang et al. (7)11 sourcesSource
PapersBuildersSelected10:24
Speculative Decoding for MoE Models: Deduplication and Draft Length

Deduplicating expert reads sounds appealing for MoE inference, but the measured bandwidth numbers show it rarely helps at practical draft lengths, which saves practitioners from engineering it blindly. Inference engineers and MoE serving teams can reuse the fitted-bandwidth method, which takes under an hour per model, to predict whether deduplication wins on their own hardware. The tested stacks — llama.cpp CUDA and Metal plus MLX-LM — cover the common Mac and NVIDIA deployment paths.

arXiv: DeepSeek@Yuzheng LanSource