CacheSpec is an inference optimization framework that transforms Program-of-Thoughts (PoT) style programs into reusable cache objects for large language models. By employing a small model for semantic variable extraction on cache hits and speculative drafting during target-LLM generation, CacheSpec reduces inference latency and improves cache reuse. Experiments on shopping, web, formula, and code QA datasets demonstrate up to 3.1× speedup in latency and 2.8× throughput gains over traditional PoT methods, while maintaining or improving task quality.
By Jingquan Chen, Jie Feng, Jinghua Piao, Shaogang Hu, Yong Li
WorldMind is a decoupled framework for state-aware NPC behavior in game world models, separating interactive world modeling into four layers: Understanding, Decision, Control, and Generation. It constructs a compact state from generated frames, reasons over it to plan NPC actions, translates actions into temporally aligned conditions, and synthesizes visual outcomes. Experiments on the newly introduced BOSS-140K dataset show that WorldMind achieves more tactically appropriate and coherent NPC behavior than baseline models in about 70% of pairwise comparisons.
By Zhiyang Deng, Boran Zhang, Danze Chen, Yeying Jin
The study examines how multilingual large language models (LLMs) produce outputs that differ across sociocultural contexts, highlighting that identity labels and source-language cues can mislead assessments of cultural grounding. Using a human‑validated, multi‑agent audit on 89,253 outputs from 12 LLMs in English, French, and Chinese across 18 occupations and three task conditions, the authors find that bias representation varies systematically by language and task. Removing direct identity cues reduces identity‑label prediction in English and Chinese but not in French, and the source language’s cultural context consistently receives the highest relevance score, though this signal weakens after translation or name masking.
"whyItMatters":"The findings show that surface cues can obscure true cross‑cultural patterns, underscoring the need for careful audit designs to avoid misleading conclusions about bias in multilingual LLMs."
By Yuanjun Feng, Tanzhou Liu, Stefan Feuerriegel, Yash Raj Shrestha
The paper introduces the Unified Path Planner (UPP), a graph‑search algorithm that balances safety and optimality by adaptively weighting heuristics and using a local inverse‑distance safety field. UPP auto‑tunes its parameters during search, guaranteeing suboptimality bounds while improving obstacle clearance. Evaluation on ten simulated environments shows UPP achieving a 0.94 OptiSafe score—significantly higher than existing methods—while adding only 0.5–1% to path length and maintaining a 100% success rate, with hardware validation on a TurtleBot confirming practical benefits.
By Jatin Kumar Arora, Soutrik Bandyopadhyay, Sunil Sulania, Shubhendu Bhasin
arXiv:2608.23104v1 Announce Type: cross
Abstract: Molecular science represents an important frontier for LLM-based agents. Unlike general agents that mainly operate over natural language, code, or we...
By Jiatong Li, Wengyu Zhang, Weida Wang, Yuxuan Ren, Wei Liu, Chenyang Mao, Yuqiang Li, Yatao Bian, Changmeng Zheng, Xiaoyong Wei, Qing Li
arXiv:2608.22830v1 Announce Type: new
Abstract: Deploying LLMs for enterprise Text-to-SQL is bottlenecked less by the model than by what context reaches it: business logic spans thousands of tables,...
By Kate Gwimm, Carson Eisenach
arXiv:2608.21969v1 Announce Type: new
Abstract: Humans have multiple levels of temporal abstractions on daily interaction and thinking, such as concept perception and strategic planning. Inspired by...
By Xiaoyu Wang, Qingqing Gu, Yue Zhao, Teng Chen, Yuqi Cao, Xiaokai Chen, Hongyan Li, Luo Ji
The paper introduces the Multi-Context Fusion Transformer (MFT), a model that predicts pedestrian crossing intentions in urban settings by integrating four types of contextual information—pedestrian behavior, environment, localization, and vehicle motion—through a progressive fusion strategy. MFT uses intra-context attention for reciprocal interactions within each context, cross-context attention to combine these contexts into a global representation, and guided attention mechanisms to refine both context tokens and the global token. Experiments on JAADbeh, JAADall, and PIE datasets show MFT outperforms existing methods with accuracies of 73%, 93%, and 90% respectively, and ablation studies confirm the importance of each network component and input context.
By Yuanzhe Li, Hang Zhong, Steffen M\"uller
Meta-Ctrl is a constrained‑decoding framework that guarantees syntactic and semantic constraints in robot plans generated by large language models. It introduces meta‑tokens—a compact vocabulary of grounded actions—to enforce syntax at the token level and semantics (preconditions, goals, ordering) at the action level, dramatically reducing memory usage from over 107 TB to under 2 GB. The approach enables a small open‑weight LM to outperform GPT‑4 on subgoal success rates in the WAH‑NL LoTa‑Bench protocol and consistently improves performance on the Embodied Agent Interface, while real‑world tabletop robot experiments show every plan satisfies its preconditions and goals by construction.
By Gwen Yidou-Weng, Edward Sun, Tianyi Ma, Metin Alp Dogan, Benjie Wang, Allen Peng, Guy Van den Broeck, Yuchen Cui
The paper introduces Delta, a two‑phase framework for testing deep reinforcement learning agents. In the first phase, the agent under test is evaluated for catastrophic failures while collecting decision‑making data. The second phase trains a challenger agent from this data using offline RL; comparing the challenger’s rewards to the original agent reveals optimality bugs, and Delta successfully uncovered thousands of such issues across multiple environments.
By Junda He, Jieke Shi, Zhou Yang, Mingfei Cheng, David Lo
VeCAS is a two‑stage framework that generates contrast‑free X‑ray angiograms from non‑contrast images. Stage I localizes vascular structures using a discriminative model and cross‑modality latent distillation, while Stage II synthesizes angiographic appearance within those regions with a vessel‑focused inpainting model. Experiments on a lower‑limb vascular intervention dataset show that VeCAS improves vascular structural fidelity and image quality, and robotic guidewire navigation tests demonstrate significant reductions in time to target and operation steps.
By De-Xing Huang, Chen-Yu Wang, Hao Liang, Xiao-Hu Zhou, Mei-Jiang Gui, Tian-Yu Xiang, Qin-Yi Zhang, Chen Wang, Xiao-Liang Xie, Shi-Qi Liu, Ming-Yuan Liu, Zhen-Chang Wang, Zeng-Guang Hou
arXiv:2608.21377v1 Announce Type: cross
Abstract: Sycophancy in large language models, the tendency to prioritize user agreement over truthful responses, has been documented extensively but studied p...
By Thantham Jittham
arXiv:2608.22176v1 Announce Type: new
Abstract: Large Language Models (LLMs) are increasingly applied to clinical prediction tasks such as in-hospital mortality and readmission from electronic health...
By Jun Hou, Yi Fang, Xuan Wang
arXiv:2608.23179v1 Announce Type: cross
Abstract: Large language model (LLM) agents are increasingly attractive for automating network configuration, yet their reliability and failure patterns are po...
By Chang Liu, Xiaohui Xie, Xinyi Chen, Yong Cui
arXiv:2608.21372v1 Announce Type: new
Abstract: This report summarizes the work conducted on the Game of Hidden Rules (GOHR), focusing on reinforcement learning agents trained to infer hidden rules f...
By Christo Mathew, Wentian Wang, Jacob Feldman, Lazaros K. Gallos, Paul B. Kantor, Vladimir Menkov, Hao Wang
arXiv:2608.21444v1 Announce Type: new
Abstract: Multi-drone systems are increasingly positioned for safety-critical missions such as search and rescue (SAR) and critical infrastructure monitoring. Ye...
By Timothy Merritt, Alejandro Jarabo-Pe\~nas, Juan Bravo-Arrabal, Maria-Theresa Bahodi, Anders Lyhne Christensen
arXiv:2608.21942v1 Announce Type: new
Abstract: Software systems have traditionally been organized around applications where human users act as principal decision-makers. Recent developments in agent...
By Mehul Goenka, Tejas Pathak, Siddharth Asthana
arXiv:2608.21976v1 Announce Type: new
Abstract: Agentic AI has automated parts of scientific discovery, including paper generation, expert-level coding, therapeutic proposal, and autonomous experimen...
By Tianyi Yu, Chengxing Tao, Haoxuan Shen, Huiyang Li, Rugang Chen, Long Teng, Lilin Wang, Yan Li, Qingbin Chen, Chaogang Xu, Lizhong Wang
arXiv:2608.22026v1 Announce Type: new
Abstract: Accurate simulation of the long-time evolution of systems governed by partial differential equations (PDEs) is central to scientific computing. Among e...
By Maqun Zhang, Feng Gao, Wankun Chen, Hui Yu, Yanhai Gan, Junyu Dong
arXiv:2608.22708v1 Announce Type: new
Abstract: Tool use in LLM systems faces a structural trade-off. Progressive disclosure keeps the prompt small by showing only the tools relevant to the current t...
By Donghui Zha, Lingwei Xu, Linxiao Wu, Yixue Dong, Haochen Li