Every processed story in chronological order, with the newest coverage first. Filter by tag, source, or score to drill in.
MiniPIC removes the requirement for identical prefixes to reuse KV cache entries, enabling efficient caching of recurring structured inputs in retrieval-augmented and agentic workloads without the large server-side code changes or host-to-device transfer overhead of prior PIC approaches.
The framework and dataset directly extend multimodal medical AI to seven major Indian languages, addressing the lack of equitable AI-driven healthcare assistance in multilingual, low-resource settings like rural India that English-centric MLLMs cannot serve.
The evidence-first protocol directly reduces the conversational bias that causes standard LLM assistants to follow misleading user hypotheses, improving diagnostic accuracy over both direct prompting and reasoning-only baselines across multiple LLM backbones.
InterleaveThinker removes the architectural barrier that has prevented existing image generators from producing interleaved text-image sequences, extending a capability previously limited to frontier models like GPT-5 to any image generator via a plug-in multi-agent pipeline.
Teams building multi-agent LLM pipelines can use behavioral economics game benchmarks as a cheap pre-screening tool to identify which open-weight models will cooperate effectively before investing in full-scale deployments.
Practitioners building multi-purpose agents can use this curriculum framework to diagnose and address capability gaps that single-domain training pipelines structurally cannot detect, such as the SACP failure mode identified in over-specialized security agents.