MASCAL trains multi-agent systems to pass calibrated confidence
Upstream agents hand off a correctness probability so downstream agents can trust, verify, or recheck web and reasoning answers.
Stories per day · 30 days
Relatedconfidence-calibration1context-compression1cross-attention1domain-adaptation1image-generation1knowledge-distillation1knowledge-transfer1latent-communication1multi-agent1multimodal-models1parametric-memory1reasoning-distillation1reinforcement-learning1retrieval1student-models1trace-selection1web-retrieval1
13 stories, newest first · one card per story
Stories tagged #question-answering
LatComm compresses LLM context once for reuse across models
On QuALITY the receiver sees 57 times less input and answers an article's questions three times faster.
Deep Fusion Memory adds chunk-level parametric memory to LLMs
With the backbone frozen, training needs 32 times fewer retrieval queries and inference uses no explicit retrieval.
ReDiST builds student-conditioned traces for reasoning distillation
Offline selection ranks traces by answer recovery and verified-answer probability before distilling only the student model.
Paper tests whether image generation expands model knowledge
Across seven knowledge categories and three unified models, visual supervision transferred unevenly into text-only question answering.
Investigating the Role of Reasoning-Language Alignment in Monolingual Retrieval-Augmented Generation
German RAG tests show reasoning-language alignment helps, but unconstrained English still matches it.
ViLegalExpert: A Large-Scale Benchmark for Vietnamese Legal Retrieval and Question Answering from Real-World Consultations
ViLegalExpert benchmarks Vietnamese legal retrieval and QA on 172K real consultation questions.
BELIEFRAG: Making Adaptive RAG State-Aware under Evolving Evidence
BELIEFRAG tracks evidence state for adaptive RAG, matching quality with 35-39% fewer tokens than iterative retrieval.
LEAP: Learned Block-wise Evidence Retrieval for Long Audio-Video Perception
LEAP retrieves short audio-visual windows for long recordings, beating Qwen3-Omni by up to 16.8%.
Asking for What Was Never Requested: Horizontal and Vertical Proactivity in Agents
Paper defines horizontal and vertical agent proactivity and trains a questioner that beats a 15x larger prompted model.
ANTMAN: Adaptive Need Tracking for Multi-Agent Navigation in Large Information Spaces
ANTMAN's Need Graph keeps multi-agent coordination nearly flat, rising 1.23x when searchable context grows 16x.
TimeEvo: Failure-Driven Self-Evolution of a Time Series Agent
TimeEvo builds a time-series agent tool library from diagnosed failures and raises accuracy on ten QA tasks.
MemoryAthena: Adaptive Routing over Latent and Generated Memories
MemoryAthena routes generated memory as a bounded correction to direct Engram retrieval.