Measures how agent memory systems miss implicitly associated facts by introducing InMind, a 125-task benchmark with paired controls that separate stored-vs-retrieval vs knowledge gaps. Quantifies a large retrieval-interface blind spot and points to routing as the core open problem.
AI Infra2026
Indexes chemistry literature as provenance-bearing atomic claims and provides a faceted taxonomy, evidence graph, and REST/SDK/MCP APIs so researchers and AI agents can retrieve verifiable, claim-level findings across papers; live index contains 2.4M claims from 147K papers.
Estimates the visually attributable portion of a privileged teacher’s next-token corrections and reconstructs student-anchored training targets for multimodal on-policy distillation. Uses counterfactual teacher queries and a signed proxy to raise supported tokens and suppress refuted ones, improving fine-grained visual knowledge transfer across model scales.
A continuous-latent diffusion language model that preserves a high-capacity, decodable text latent and directly models its distribution via a block-causal diffusion transformer and query-based encoder–decoder; achieves top results on OpenWebText and XSum while scaling to 1B parameters.
Evaluates how large language models fabricate user attributes in personalization and whether model self-monitoring is a reliable signal. Introduces MirageBench (150 personas, 6 personalization tasks, judge-validated faithfulness taxonomy) and a 12-model leaderboard revealing pervasive over-inference and a 'Self-Monitoring Inversion'.
Yingpeng Ma, Jianhao Yan +7Affiliation: NLP2CT Lab, University of Macau, Macau, China, Affiliation: Westlake University, Hangzhou, China +3
Evaluates whether LLM-driven storyteller agents preserve long-horizon logical consistency under adversarial player interventions. Introduces NCP-Bench (100 movie-derived narrative environments) with structured trajectory/commitments and automatic violation checks; finds strong LLMs often contradict themselves across multi-turn interactions.
Zhuoyang Qian, Biao Wu +7Affiliation: Vast Intelligence Lab
University of Technology Sydney, Affiliation: Equal Contribution
Corresponding Authorhttps://github.com/Spark-To-Paper-Skills/[email protected]
Generates full publication-format research papers from a short idea by composing 13 coding-assistant skills; it retrieves literature, plans and runs feasible experiments, produces editable vector figures, and enforces deterministic integrity checks so claims are revised to match measured evidence.
AI Agent Papers2026
Uses a stronger 'builder' model at inference time to construct executable harnesses that boost weaker target models without parameter updates, mainly by turning unstable reasoning into deterministic code, routing, and strict answer-format enforcement.
Generates unified embeddings for text, images, video, visual documents and interleaved multimodal inputs with configurable output dimensions and Matryoshka truncation to trade accuracy for cost. Model weights and code are released under Apache-2.0; the 9B variant scores 80.6 on MMEB-v2.
AI Agent Papers2026
Analyzes how to generate useful interaction data for LLM agents and proposes the ACE lens — Accuracy, Complexity, divErsity — while factorizing agentic data as (E, q, τ, v). Surveys verification, difficulty calibration, and coverage strategies and outlines implications for training and benchmarks.
Analyzes on-policy distillation for LLM fine-tuning, shows teacher token-level supervision is often noisy and not the main driver of gains, and introduces OPSA, a supervision-free, entropy-adaptive method that suppresses low-probability tokens to improve downstream accuracy.
Turns sparse per-student records into individualized simulators that both reproduce a student’s responses and update them under tutor guidance using pooled LLM pretraining followed by per-student specialization; releases StudentSimEval and reference simulators across chess, L2 writing, and math.