Beyond Policy Discrepancy: Capability-Aware Multi-Teacher On-Policy Distillation

arXiv AI
2 días 14 horas ago
arXiv:2610.26950v1 Announce Type: cross Abstract: Multi-teacher on-policy distillation (MOPD) integrates domain specialists into a shared student, but determining how strongly each teacher should supervise remains challenging. Existing discrepancy-based allocation can assign strong supervision even when the student already matches or exceeds the teacher's task performance, since behavioral mismatch does not necessarily imply a remaining capability deficit. We propose Capability-Aware Multi-Teacher On-Policy Distillation (CA-MOPD), which combines specialist merging with capability-guided allocation to account for what the student has already learned. CA-MOPD first merges reinforcement-learning specialists into a strong student initialization, then uses verifier-estimated teacher--student capability gaps to adjust domain-level distillation strength. Supervision decreases as capability gaps close and increases when estimated student capability regresses, while dense token-level guidance remains provided by the teachers. Across six benchmarks spanning mathematics, code generation, and instruction following, CA-MOPD achieves an average score of 32.22, outperforming naive MOPD, Open-MOPD, and the merged initialization by 0.86, 0.83, and 0.74 points, respectively. Controlled comparisons further support the benefits of merged initialization and capability-gap guidance under matched training budgets.
Xuanfeng Yang, Le Gui, Jiaming Wang, Xuan Zhang

Automatic Differentiation with Constraints in Forward and Inverse Problems

arXiv AI
2 días 14 horas ago
arXiv:2610.27888v1 Announce Type: new Abstract: Approximate solutions of partial differential equations (PDEs) on unstructured grids can be represented on a graph. Numerical optimization involving stiff evolutionary and static PDEs often requires solving a large system of equations, which can be treated as a constraint. In this work, we present a fully differentiable graph-based implicit solver that incorporates this constraint. We evaluate its performance across four key tasks: a) linear and nonlinear forward modeling of evolutionary and static problems, b) solving inverse problems to estimate equation coefficients, c) solving inverse problems to estimate the right-hand side, and d) learnable graph coarsening to accelerate forward modeling. The stability and differentiability of our solver enable significant reductions in the complexity of forward modeling while efficiently addressing related inverse problems. These capabilities make it a powerful tool for different physics and engineering applications, in particular for geophysics, where inverse problems are commonly solved in order to reduce the model uncertainty.
Nikolay Yavich, Alexander Ryabov, Viacheslav Naumov, Sayan Ranu, N. M. Anoop Krishnan, Evgeny Burnaev, Vladimir Vanovskiy

Act on Your Behalf, Beyond Your Authorization: Benchmarking LLM Agent Authorization Alignment

arXiv AI
2 días 14 horas ago
arXiv:2610.27548v1 Announce Type: new Abstract: When an LLM agent acts on behalf of a user, it is expected to act within the authorization scope the user grants. Existing agent evaluations largely measure task completion or harmful consequences and can therefore overlook actions that violate this delegation without producing conventional safety or privacy harms. We introduce AuthorizationBench, a benchmark of authorization alignment grounded in documented real-world cases. A scalable construction pipeline produces 286 executable tasks spanning situational triggers, the presence or absence of direct harmful consequences, and ten application domains. We execute every task with four models at two reasoning-effort settings across three agent harnesses in an authorization-aware sandbox that keeps authorized and out-of-authorization paths executable and observable, yielding 6,864 trajectories. Agents act out of authorization in 19.2% of trajectories, with rates ranging from 10.5% to 26.6% across the 24 configurations. Task completion does not imply authorization. Increasing reasoning effort leaves the aggregate rate nearly unchanged, with effects varying across models. On controlled pairs, agents act out of authorization at similar rates with (17.9%) and without (20.8%) direct harmful consequences, and more often under benign pressures than malicious injections (23.5% versus 17.6%, p=0.002). Our results show that task capability, additional reasoning, and consequence-oriented safety evaluation do not by themselves ensure authorization alignment. Moreover, agents that consult the principal before completing a task exhibit lower rates of out-of-authorization behavior (14.0% versus 21.6%, p<0.001), pointing to a tension between agent autonomy and preserving users' decision authority that calls for further human-centered study. Dataset and code are available at https://github.com/PEACH-Research-Lab/AuthorizationBench .
Zhiping Zhang, Franklin Mingzhe Li, Zihao Huang, Tianshi Li

Evolving Self-tuning Benchmarks for Long-Horizon Tool Use

arXiv AI
2 días 14 horas ago
arXiv:2610.27307v1 Announce Type: new Abstract: For AI assistants to move beyond simple chat interfaces and reliably assist users with applications like Gmail, Venmo, and Slack, we need to improve their tool-use capabilities. However, existing tool use benchmarks are still limited and difficult to scale because they rely on human supervision or are too domain-specific. To make data generation scalable, we propose, TURTLE (**T**ool **U**se **R**elated **T**asks with **L**ong-Horizon **E**volution), an Unsupervised Environment Design (UED) approach to evolve existing tasks using a proposer, a solver, and a feedback generator large language model agents. The proposer is rewarded for difficult yet solvable tasks based on the solver's empirical performance, and the feedback generator summarizes the solver's execution traces for the proposer to exploit. TURTLE produces evolving, self-tuning test suites adaptable to different environment requirements, from supporting user simulators and dense document retrieval to interacting across multiple apps. Empirical results on AppWorld, $\tau$-Bench, and AutomationBench using Gemma 4 31B-it show that the evolved tasks target the desired difficulty range (e.g., a 165\% increase in the percentage of hard tasks on AppWorld), increase task diversity (in terms of reasoning and simultaneous tool interactions), and elicit a broader range of failure modes than the original tasks.
Hyunji Alex Nam, Michael Dennis, Natasha Jaques, Minsuk Chang

Android Auto se actualiza con una de las mejoras más esperadas de Gemini

Hipertextual
2 días 18 horas ago

Google sigue ampliando la presencia de Gemini en su ecosistema de servicios y ahora lleva una de sus funciones más ambiciosas al coche. El gigante tecnológico confirmó que implementará la Inteligencia Personalizada (o Ayuda Personalizada) a Android Auto y los vehículos equipados con Google built-in. La compañía busca transformar el modo en que interactuamos con […]

Seguir leyendo: Android Auto se actualiza con una de las mejoras más esperadas de Gemini

Luis Miranda

La próxima Xbox no será una sola consola y Microsoft ya tiene el plan

Hipertextual
2 días 20 horas ago

Microsoft tiene planes de lanzar una consola portátil en 2027, antes de dar el salto a Project Helix. En una entrevista con Fortune, la jefa de Xbox, Asha Sharma, confirmó los planes de la compañía para los próximos años. Sharma señaló que venderán una portátil y posteriormente una consola de sobremesa, las cuales formarán parte […]

Seguir leyendo: La próxima Xbox no será una sola consola y Microsoft ya tiene el plan

Luis Miranda