Ive come up with an adversarial RL design that could potentially push LLMs to superhuman level reasoning in a variety of domains. The setup would involve 3 actors. First is the problem generator. Its tasked to simply generate a problem and solution lets say for coding. Second is the validator agent. this agent is frozen, all it does is take the problem generated by the solver and then asks some important questions like, “is the problem syntactically correct?” […]

Ver mais

arXiv:2603.25958v1 Announce Type: new Abstract: The Minkowski weighted k-means (mwk-means) algorithm extends classical k-means by incorporating feature weights and a Minkowski distance. Despite its empirical success, its theoretical properties remain insufficiently understood. We show that the mwk-means objective can be expressed as a power-mean aggregation of within-cluster dispersions, with the order determined by the Minkowski exponent p. This formulation reveals how p controls the transition between selective and uniform use of features. Using this representation, we derive bounds […]

Ver mais

Large language models (LLMs) have recently demonstrated impressive performance on complex, multi-step reasoning tasks, especially when post-trained with outcome-rewarded reinforcement learning Guo et al. 2025. However, it has been observed that outcome rewards often overlook flawed intermediate steps, leading to unreliable reasoning steps even when final answers are correct. To address this unreliable reasoning, we propose PRoSFI (Process Reward over Structured Formal Intermediates), a novel reward method that enhances reasoning reliability without compromising accuracy. Instead of generating formal […]

Ver mais

Catastrophic forgetting, the tendency of neural networks to forget previously learned knowledge when learning new tasks, has been a major challenge in continual learning (CL). To tackle this challenge, CL methods have been proposed and shown to reduce forgetting. Furthermore, CL models deployed in mission-critical settings can benefit from uncertainty awareness by calibrating their predictions to reliably assess their confidences. However, existing uncertainty-aware continual learning methods suffer from high computational overhead and incompatibility with mainstream replay methods. To […]

Ver mais

arXiv:2602.15898v1 Announce Type: new Abstract: Multi-hop question answering (QA) necessitates multi-step reasoning and retrieval across interconnected subjects, attributes, and relations. Existing retrieval-augmented generation (RAG) methods struggle to capture these structural semantics accurately, resulting in suboptimal performance. Graph-based RAGs structure such information in graphs, but the resulting graphs are often noisy and computationally expensive. Moreover, most methods rely on single-step retrieval, neglecting the need for multi-hop reasoning processes. Recent training-based approaches attempt to incentivize the large language models (LLMs) […]

Ver mais

arXiv:2601.17855v1 Announce Type: cross Abstract: Load balancing-the allocation of work across parallel resources to reduce delay, energy and cost-is a pervasive challenge in science and engineering, from large-scale simulation and data processing to cloud and manufacturing operations. Motivated by the emerging bottleneck in large language model (LLM) serving, we study a particularly stringent regime of load balancing that arises in barrier-synchronized, stateful systems: work cannot be freely migrated and progress is gated by the slowest participant at each […]

Ver mais