Wei Wang

Pengfei Zhou, Zhiwei Tang, Yixing Ma et al.

Real-world users typically have access to multiple Large Language Models (LLMs) from different providers, and these LLMs often excel at distinct domains, yet none dominate all. Consequently, routing each task to the most suitable model becomes critical for both performance and cost. Existing routers treat this as a static, one-off classification problem. However, we identify the performance bottleneck for these routers as information deficit: simply augmenting a vanilla LLM router with performance statistics at the task-dimension level yields a 15.3% relative gain, surpassing a heuristic router built on the same dimension-level priors. Motivated by this finding, we propose Agent-as-a-Router, a framework that formalizes routing as a C-A-F loop (Context->Action->Feedback->Context). It closes the information gap by accumulating execution-grounded experience during deployment. We instantiate this framework as ACRouter, composed of an Orchestrator, a Verifier, a Memory module, and introduce CodeRouterBench, an evaluation environment comprising ~10K task instances with verified scores from 8 frontier LLMs, enabling regret-based router comparison on streaming tasks. Experiments show that ACRouter achieves the lowest cumulative regret on in-distribution tasks and generalizes to out-of-distribution agentic-programming tasks, demonstrating that our routing framework actively closes the information gap. Codes and benchmarks are released at https://github.com/LanceZPF/agent-as-a-router.

Jin-Guo Liu, Shang-Qi Lu, Xin-Ran Shi et al.

This paper is an experience report on a 13-week Test-Driven, AI-Assisted (TDAA) redesign of DSAA 3071, Theory of Computation, an upper-level course at the Hong Kong University of Science and Technology (Guangzhou). The design is simple: the course replaces lectures with self-directed, AI-assisted learning, and frequent, independently completed tests create a high-frequency quality gate. AI agents help the instructor prepare the learning path, course website, tests, grading workflow, and repairs. Two conditions made this strict gate workable. Students needed a visible preparation path of learning sheets and aligned validation practice, so the closed-book tests felt fair rather than arbitrary. The instructor needed an AI-assisted materials harness, a version-controlled agent workspace, so that weekly drafting, review, test production, and grading could scale with human oversight. Evidence from a student survey ($N=18$), weekly scores, and the project's git history suggests that students treated the tests as useful accountability and that the harness made frequent closed-book testing operational. The evidence is limited to one small, proof-heavy course without a control group. The contribution is therefore a reusable design pattern: high-frequency tests preserve individual accountability, while AI agents make material production and marking scalable. We release the harness as a public starter template so that other instructors can reproduce it.

Masao Someki, Alexander Polok, Carlos Carvalho et al.

Recent speech research involves increasingly large datasets, complex models, and diverse experimental workflows. However, existing frameworks require substantial engineering effort to support such experiments. We present ESPnet3, a speech and audio research framework built on a modular system architecture with configuration-driven dataset composition and unified Python-based workflows. ESPnet3 introduces a DataOrganizer abstraction for flexible dataset integration and dataset sharding for memory-efficient large-scale training, while allowing recipe-specific logic through lightweight stage overrides. In OWSM pre-training experiments, ESPnet3 reduces per-epoch training time by \emph{21.1 minutes} compared to ESPnet2 and achieves \emph{>80\% GPU utilization} in multi-node training. Fine-tuning experiments show that new models and datasets can be integrated with around \emph{46 lines of additional code}. ESPnet3 will be publicly released with model checkpoints and training logs.

Jack Stonecipher, Allen Gao, Wei Wang

Fish achieve efficient swimming across varied speeds through active modulation of their body flexibility. To explore the effects of tunable stiffness on swimming performance, we present a bio-inspired freely swimming fish robot with a rapidly tunable particle-jamming body. This design enables rapid stiffness adjustments with negligible changes in shape or volume, achieving a 54% variation in flexural rigidity across vacuum pressures of 0 to -40 kPa. We visualize the midline of the oscillating body under both low- and high-stiffness conditions, and the comparison confirms that the body curvature varies with stiffness. We further experimentally evaluate the tunable stiffness body's effects on swimming performance using velocity and cost of transport (CoT) measurements obtained via a motion tracking system. Results show that active stiffness tuning is essential for sustaining efficient and high-speed swimming across beating frequencies of 1-3 Hz. At low frequencies (1-1.5 Hz), a softer body (0 kPa) maximizes velocity and minimizes CoT, whereas at high frequencies (2.5-3 Hz), a stiffer body (-40 kPa) delivers superior velocity and reduced transport cost. These findings highlight stiffness modulation as a key strategy for adaptive and efficient propulsion in bio-inspired robotic swimmers.