Large language models often apply complex reasoning indiscriminately, leading to overthinking on simple tasks and underthinking on harder ones. We present Route-and-Generate (Rouge), a hierarchical training framework that enables LLMs to adaptively decide when and how much reasoning is required for each input. Our appr...
Tien-Phat Nguyen, Truong Nguyen, Thanh-Thien Le et al.· 0 citations
It is concluded that consistency under reordering can be optimized without improving the moral judgment it is meant to signal: consistency under reordering can be optimized without improving the moral judgment it is meant to signal.
Marius De Kuthy Meurers, Aaron T. Parisi, Bernd Bohnet· 0 citations
End-to-end autonomous driving has achieved state-of-the-art performance on benchmarks and real-world deployments. Its standard training recipe, however, is expensive across all existing paradigms: imitation learning requires fleet-scale driving logs whose behavior distribution only thinly covers the safety-critical tai...
Zikang Xiong, Wei-Xin Li, Zhouchonghao Wu et al.· 0 citations
Supervised fine-tuning adapts language models to specialized tasks but can disrupt previously acquired behavior. Existing approaches constrain parameters, gradients, or replay objectives, yet the connection between the protected object, the intervening trajectory, and retained outputs remains unclear. We develop a func...
Qingsen Ma, Dianyun Wang, Jiahong Liu et al.· 0 citations
Reach audiences
Advertise in front of researchers, engineers, and readers.
Program optimization aims to improve performance while preserving behavior. Coding agents can already make programs faster by changing algorithms, data structures, and execution strategies. However, existing equivalence checks mostly rely on tests, which agents have been shown to exploit, passing the tests without pres...
Zhe Ye, Udaya Ghai, Zhizhen Qin et al.· 0 citations
A method to detect synchronous model training in facility-level power by Viterbi tracking of the wandering iteration cadence and a modest passive meter is sufficient to make hidden training both detectable and expensive, which makes power metering a promising part of compute verification that does not depend on trustin...
T. Kimpson, Mauricio Baker, Emlyn Graham· 0 citations
Optimizing Information Retrieval (IR) pipelines for Retrieval-Augmented Generation (RAG) requires jointly selecting structural components and numerical hyperparameters under limited evaluation budgets. We introduce the Information Retrieval Agentic Harness (IRAH), which combines Language Agent Tree Search for IR (LATS-...
Maria Ana Cardei, Aniket Singh, Amir Moghadam et al.· 0 citations
Every interaction between AI agents begins with a handshake: the initiating agent must locate its counterpart, learn what it can do, and establish a session before the first useful message can be sent. Surveys of agent communication protocols name handshake overhead as an efficiency criterion, yet none report measureme...
Wrong automatic routing of an IT support ticket can cost more than human triage. We evaluate cost-aware abstention as an offline, cost-conditional readiness assessment: select a routing policy on calibration data under explicit costs, then evaluate that frozen policy once on held-out test data. On a public 47,837-ticke...
Meta-reinforcement learning aims to train agents that adapt quickly to unseen tasks from limited interaction, but this requires representing task evidence over long interaction histories. Classical recurrent agents compress experience into hidden states, while transformer-based agents retrieve from past trajectories at...
Ali Najar, Hosein Hasani, Amir Kooshan Fattah et al.· 0 citations
Chain-of-thought (CoT) prompting elicits step-by-step rationales that appear to explain model predictions, but there is no guarantee these steps causally drive the final answer. In standard autoregressive generation, dense attention allows each reasoning step to attend to the full context, enabling the model to derive...
Gabriele Dominici, T. McMillan, Fazl Barez et al.· 0 citations
Instrumental convergence is often invoked as a general reason to expect capable agents to acquire or express power-seeking subgoals: self-preservation, resource acquisition, deception, or resistance to oversight which poses an important threat to the society. For LLMs, reinforcement learning is the first stage of post-...
