DentAgent is an evidence‑centric multi‑agent framework designed for multimodal dental reasoning. It coordinates five specialized agents that process domain knowledge, radiographs, intraoral photographs, and 3D dental data, converting observations into structured evidence records. The Evidence Blackboard tracks coverage, gaps, and conflicts before generating responses, and the system outperforms senior specialists by 17.3 percentage points on multi‑label diagnosis across four benchmarks.
By Zijie Meng, Xiwei Dai, Yixuan Tang, Jin Hao, Yang Feng, Fudong Zhu, Xiaoqiang Liu, Shaosheng Cao, Zuozhu Liu
The paper introduces an adaptive memory and reflection (AMR) multi‑agent system for medical question answering. Each agent has dedicated memory and uses reflection‑based feedback to retrieve relevant prior cases, improving reasoning. The system routes questions through solo, collaborative, or escalated workflows and includes consensus and ethical overseer modules, achieving strong performance on MedQA and MedMCQA datasets.
By Pradeep Murugesan, Luoxiao Yang, Xueli Chen, Xinqi Fan
The paper introduces VAKE, a two‑stage reinforcement‑learning framework that activates latent factual knowledge in large language models. In the Priming stage, the model explicitly inserts bridging triples into an insufficient subgraph, guided by rewards from a frozen model’s answers. The Reasoning stage then trains the model to answer from the original input, demonstrating that the elicitation capability transfers to implicit reasoning and consistently outperforms baselines across multiple benchmarks and model sizes.
By Zuocheng Ying, Yang Yang, Yumou Wu, Chuanbo Zhu, Jiarui Wang, Ziqi Wu, Jingming Cai, Junqing Yu, Zikai Song
The paper tackles the challenge of reconstructing argument graphs from natural language by addressing enthymemes—arguments with implicit premises. It proposes a neuro‑symbolic pipeline that employs large language models to generate intermediate implicit premises, translates them into logical formulas, and combines them with explicit premises and claims to determine entailment, contradiction, or neutrality. The method is evaluated on the Microtext Argumentative Corpus.
By Xuyao Feng, Anthony Hunter
The paper proposes a systematic framework for creating a "Map of Datasets in Engineering Design and Systems Engineering" (EDSE) to address the fragmented and inaccessible nature of existing datasets. It introduces a multi‑dimensional taxonomy that classifies datasets by domain, lifecycle stage, data type, and format, and presents an interactive discovery tool built on a knowledge graph data model. The authors analyze the current data landscape, identify underrepresented areas such as early‑stage design and system architecture, and suggest strategies for curation and sustainability to build a dynamic, community‑driven resource.
By H. Sinan Bank, Daniel R. Herber
MedStruct‑S is a benchmark for semi‑structured information extraction from OCR‑derived clinical reports, covering key discovery, key‑conditioned QA, and end‑to‑end key‑value extraction. It contains 3,582 annotated real‑world report pages and evaluates models under unknown keys and OCR noise. Experiments show encoder‑only models excel at non‑null key‑conditioned QA, while fine‑tuned decoder‑only models achieve the strongest overall performance across model sizes.
By Yingyun Li, Yu Wang, Haiyang Qian
DeepWeaver addresses the evidence synthesis gap in open‑ended question answering by weaving noisy retrieved evidence into comprehensive answers. It introduces Thought Block Chains (TBCs) that organize claims, key information, and citations, allowing the system to revise and expand evidence before final generation. Evaluations on LoQA and DeepResearch Bench show improved content sufficiency, citation grounding, and detail preservation across multiple LLMs.
The paper introduces TextQ‑German, a dataset suite for evaluating German natural language generation (NLG) from a Quality of Experience (QoE) perspective, covering tasks such as summarization and machine translation. Human ratings collected via crowdsourcing identify perceptual quality dimensions, and the authors develop automatic QoE prediction models—including transformer‑based, linguistic feature‑based, and hybrid approaches—showing that hybrid models outperform pure transformers and that linguistic features alone can rival fine‑tuned language models. The dataset is further enriched with large language model (LLM) outputs annotated with overall QoE scores, and validation on held‑out data demonstrates generalization to unseen data.
"whyItMatters":"The resource provides a publicly accessible benchmark and baseline models for human‑centered NLG evaluation, enabling the development of systems that better align with human quality perception."
SIDScope is a diagnostic tool that evaluates Semantic-ID mappings used between item tokenizers and generative recommenders. It normalizes artifacts, verifies provenance, profiles mapping structure, compares revisions, and tracks path-to-item outcomes in generated traces. Using data from Amazon and Yelp, SIDScope shows that interface health depends on multiple signals, revealing gaps in prefix alignment, trace accounting, and refresh handling that affect model reuse.
TranslatePsy-AfriSLM is an open‑source machine‑translation resource set for 19 Sub‑Saharan African languages, comprising curated parallel data, African‑specialized synthetic data, and a suite of fine‑tuned small language models (SLMs). The authors demonstrate that a unified quality‑estimation filtering can discard up to 96% of training tokens while preserving quality, and that filtered synthetic data dominates the quality‑efficiency Pareto frontier. Models trained on this curated mixture outperform larger systems such as TranslateGemma‑27B and Qwen3.5‑122B‑A10B, achieving superior performance with as few as 0.8 B parameters.
PXDepth is a monocular depth estimation model that separates global context modeling from pixel-level depth prediction. It uses a large-patch Vision Transformer to capture scene context and a pixel-space predictor with Context‑Modulated Pixel Transformer blocks to preserve high‑resolution spatial details. The approach maintains fine structures and sharp boundaries while achieving competitive global depth accuracy in zero‑shot benchmarks.
By Zhiyuan Yuan, Guanying Chen, Lingteng Qiu, Ruimao Zhang, Shuguang Cui, Xiaochun Cao
WIP: LLM Odyssey is an open‑source, browser‑based serious gaming platform that offers 13 interactive games to teach Large Language Model engineering concepts such as tokenization, transformer architecture, prompt engineering, RAG, and production deployment. The platform is organized into three learning tiers—Cognitive Core, Systems Forge, and Foundry Arena—aligned with Bloom’s revised taxonomy, and each game employs five pedagogical strategies including immediate feedback, scaffolded hints, progressive difficulty, worked examples, and authentic production scenarios. An initial deployment at a Canadian college in Winter 2026 confirmed functional requirements, highlighted the need for adaptive difficulty, and led to the design of a formal mixed‑methods evaluation protocol for future studies.
By Priyamvada Tripathi
TokEval is a tokenizer evaluation suite that introduces metrics beyond traditional fertility and compression rate, capturing linguistically and structurally meaningful properties such as UTF‑8 character boundary integrity and digit place‑value alignment for mathematics. The authors validate these metrics by pretraining language models with varied tokenizers and measuring downstream performance on bits‑per‑byte and benchmarks covering linguistic understanding, mathematical reasoning, and code generation. Their results show that information‑theoretic metrics predict language modeling performance, while structure‑sensitive metrics correlate with task accuracy, suggesting TokEval can guide tokenizer selection more principledly.
By Clara Meister
MITRE‑SAGE is a multi‑agent retrieval‑augmented generation framework that combines semantic and structural cybersecurity knowledge to enhance large language model question‑answering. It decomposes tasks into query interpretation, evidence retrieval, and answer synthesis, supporting vulnerability assessment, threat profiling, and relationship extraction. Experiments show that MITRE‑SAGE outperforms standalone LLMs and conventional RAG methods, with a lightweight Qwen2.5‑based configuration excelling on most benchmark tasks.
By Ali Habibzadeh, Farid Feyzi, Reza Ebrahimi Atani
MoRA is a human‑centric geospatial representation learning framework that uses a large mobility graph as its backbone to fuse spatial tokenization, graph neural networks, and asymmetric contrastive learning. It aligns over 100 million points of interest, massive remote sensing imagery, and structured demographic data with a billion‑edge mobility graph, producing compact 128‑dimensional embeddings that capture socio‑economic context and functional roles of locations. On a benchmark of nine downstream social and economic prediction tasks, MoRA outperforms state‑of‑the‑art models by an average of 12.9% and demonstrates scaling behavior analogous to large language models.
By Ya Wen, Jixuan Cai, Qiyao Ma, Linyan Li, Xinhua Chen, Chris Webster, Yulun Zhou
The paper explores energy-aware knowledge distillation for large language models (LLMs) used in software engineering tasks such as clone detection, vulnerability prediction, and code summarization. It shows that the commonly used FLOPs metric does not reliably reflect actual energy consumption, and that using energy-surrogate models during distillation can reduce inference energy by up to 90% and memory usage by 86% with only modest accuracy loss. The study demonstrates that guiding distillation with direct energy estimates improves the sustainability and deployability of LLMs on consumer hardware.
By Enrique Barba Roque, Lu\'is Cruz, Annibale Panichella
The paper introduces the Structure-Internalized Rule Language Model (SIRLM) to improve Knowledge Graph Reasoning (KGR) by addressing the mismatch between KG structural context and Large Language Model (LLM) parametric knowledge. SIRLM centers on a Structure-Internalized Rule Generator (SIRG) that uses in-context learning, a structural relation memory, a KG tokenizer, and a neuro-symbolic reasoner to generate structural rules and provide faithful rule-execution feedback. Experiments on 36 datasets against 17 state‑of‑the‑art KGR methods show that SIRLM achieves significant performance gains.
By Xingrui Zhuo, Jiapu Wang, Manzong Huang, Gongqing Wu, Xindong Wu
The paper investigates how Engram-style hashed memory can be transferred between different language model backbones. By freezing a memory table trained on a source model and attaching it to a target model with only a lightweight reader, the authors find that both the memory content and correct addressing are important, but the reader must be aligned to the target to make the memory useful. In question‑answering experiments, a dual‑layer, four‑branch reader nearly matches same‑model performance, and when the reader interface is directly compatible, the frozen memory alone provides substantial benefit, with optional reader adaptation offering further gains.
By Mingyuan Li, Guangsheng Yu, Xu Wang, Shaoxiong Ji
The paper investigates why adaptive optimizers like Adam outperform SGD when fine‑tuning Transformers. It introduces gradient heterogeneity—the variation in gradient norms across parameter blocks—and shows, both theoretically and experimentally, that this heterogeneity, together with Hessian heterogeneity, hampers SGD convergence while sign‑based methods such as SignSGD are less affected. The study links the source of gradient heterogeneity to layer‑normalization placement, finding that Post‑LN architectures exhibit the strongest effect, and uses SignSGD as a tractable proxy to analyze Adam‑like behavior and learning‑rate scaling.
By Akiyoshi Tomihari, Issei Sato
The paper explains why GPT‑style language models fail to transfer directly to symbolic music. It argues that success in language comes from tokenization that compresses data by creating a coordinate system where recurring patterns become predictable. For music, the authors propose that tokenization must build a predictively effective, relationally lossless coordinate system—defining Fact–Token and Token–State boundaries—to enable compression without sacrificing contextual freedom. Controlled experiments confirm that proper coordinate construction improves predictive compressibility, whereas mere sequence compaction does not.
By Yi Wang