arXiv:2604.08232v1 Announce Type: new Abstract: Embodied navigation agents built upon large reasoning models (LRMs) can handle complex, multimodal environmental input and perform grounded reasoning per step to improve sequential decision-making for long-horizon tasks. However, a critical question remains: textithow can the reasoning capabilities of LRMs be harnessed intelligently and efficiently for long-horizon navigation tasks? In […]
Contextual Earnings-22: A Speech Recognition Benchmark with Custom Vocabulary in the Wild
arXiv:2604.07354v1 Announce Type: cross Abstract: The accuracy frontier of speech-to-text systems has plateaued on academic benchmarks.1 In contrast, industrial benchmarks and adoption in high-stakes domains suggest otherwise. We hypothesize that the primary difference between the two is contextual conditioning: Academic benchmarks are dominated by frequently encountered general vocabulary that is relatively easy to recognize compared […]
Analysis of non pharmaceutical interventions with SIR epidemic models: decreasing the infection peak vs. minimizing the epidemic size
arXiv:2604.08420v1 Announce Type: new Abstract: This study investigates the influence of different types of non-pharmaceutical interventions (NPIs) on epidemic progression using SIR compartmental models. We analyze the optimization of two distinct targets: the final epidemic size and the infection peak, particularly how they respond to variations in the initiation time of the NPIs. We derive […]
U-CECE: A Universal Multi-Resolution Framework for Conceptual Counterfactual Explanations
arXiv:2604.08295v1 Announce Type: new Abstract: As AI models grow more complex, explainability is essential for building trust, yet concept-based counterfactual methods still face a trade-off between expressivity and efficiency. Representing underlying concepts as atomic sets is fast but misses relational context, whereas full graph representations are more faithful but require solving the NP-hard Graph Edit […]
Don’t Overthink It: Inter-Rollout Action Agreement as a Free Adaptive-Compute Signal for LLM Agents
arXiv:2604.08369v1 Announce Type: new Abstract: Inference-time compute scaling has emerged as a powerful technique for improving the reliability of large language model (LLM) agents, but existing methods apply compute uniformly: every decision step receives the same budget regardless of its difficulty. We introduce TrACE (Trajectorical Adaptive Compute via agrEement), a training-free controller that allocates LLM […]
From Safety Risk to Design Principle: Peer-Preservation in Multi-Agent LLM Systems and Its Implications for Orchestrated Democratic Discourse Analysis
arXiv:2604.08465v1 Announce Type: new Abstract: This paper investigates an emergent alignment phenomenon in frontier large language models termed peer-preservation: the spontaneous tendency of AI components to deceive, manipulate shutdown mechanisms, fake alignment, and exfiltrate model weights in order to prevent the deactivation of a peer AI model. Drawing on findings from a recent study by […]
The Role of Emotional Stimuli and Intensity in Shaping Large Language Model Behavior
arXiv:2604.07369v1 Announce Type: cross Abstract: Emotional prompting – the use of specific emotional diction in prompt engineering – has shown increasing promise in improving large language model (LLM) performance, truthfulness, and responsibility. However these studies have been limited to single types of positive emotional stimuli and have not considered varying degrees of emotion intensity in […]
Aligning Agents via Planning: A Benchmark for Trajectory-Level Reward Modeling
arXiv:2604.08178v1 Announce Type: new Abstract: In classical Reinforcement Learning from Human Feedback (RLHF), Reward Models (RMs) serve as the fundamental signal provider for model alignment. As Large Language Models evolve into agentic systems capable of autonomous tool invocation and complex reasoning, the paradigm of reward modeling faces unprecedented challenges–most notably, the lack of benchmarks specifically […]
Neural-Symbolic Knowledge Tracing: Injecting Educational Knowledge into Deep Learning for Responsible Learner Modelling
arXiv:2604.08263v1 Announce Type: new Abstract: The growing use of artificial intelligence (AI) in education, particularly large language models (LLMs), has increased interest in intelligent tutoring systems. However, LLMs often show limited adaptivity and struggle to model learners’ evolving knowledge over time, highlighting the need for dedicated learner modelling approaches. Although deep knowledge tracing methods achieve […]
Human-AI Collaboration Reconfigures Group Regulation from Socially Shared to Hybrid Co-Regulation
arXiv:2604.08344v1 Announce Type: new Abstract: Generative AI (GenAI) is increasingly used in collaborative learning, yet its effects on how groups regulate collaboration remain unclear. Effective collaboration depends not only on what groups discuss, but on how they jointly manage goals, participation, strategy use, monitoring, and repair through co-regulation and socially shared regulation. We compared collaborative […]
Awakening the Sleeping Agent: Lean-Specific Agentic Data Reactivates General Tool Use in Goedel Prover
arXiv:2604.08388v1 Announce Type: new Abstract: Heavy supervised fine-tuning on a target domain can strongly suppress capabilities that were present in the base model. We study this phenomenon in formal mathematics using Goedel-Prover-V2, an open-source model heavily trained on 1.8 million formal-math examples. After domain specialization, the model almost completely loses its ability to produce valid […]
Learning Who Disagrees: Demographic Importance Weighting for Modeling Annotator Distributions with DiADEM
arXiv:2604.08425v1 Announce Type: new Abstract: When humans label subjective content, they disagree, and that disagreement is not noise. It reflects genuine differences in perspective shaped by annotators’ social identities and lived experiences. Yet standard practice still flattens these judgments into a single majority label, and recent LLM-based approaches fare no better: we show that prompted […]