arXiv:2602.04009v1 Announce Type: new Abstract: Prompt-guided generative AI models have rapidly expanded across vision and language domains, producing realistic and diverse outputs from textual inputs. The growing variety of such models, trained with different data and architectures, calls for principled methods to identify which types of prompts lead to distinct model behaviors. In this work, we propose PromptSplit, a kernel-based framework for detecting and analyzing prompt-dependent disagreement between generative models. For each compared model pair, PromptSplit constructs a […]

Ver mais

arXiv:2601.14279v1 Announce Type: new Abstract: We investigate learned KV cache compression through Speculative Importance Prediction (SIP), a 1.7M parameter non-query-aware scorer that predicts token importance from KV representations alone. Despite architectural sophistication (multi-horizon lookahead, cross-attention), SIP does not outperform simple baselines, including random selection, across 5 seeds, 4 retention levels, and 3 tasks. Key findings: (1) position-based heuristics (keep first 4 + last N tokens) match or exceed learned approaches; (2) prefill attention provides equivalent signal to complex […]

Ver mais

arXiv:2601.16282v1 Announce Type: new Abstract: Contemporary automated scientific discovery has focused on agents for generating scientific experiments, while systems that perform higher-level scientific activities such as theory building remain underexplored. In this work, we formulate the problem of synthesizing theories consisting of qualitative and quantitative laws from large corpora of scientific literature. We study theory generation at scale, using 13.7k source papers to synthesize 2.9k theories, examining how generation using literature-grounding versus parametric knowledge, and accuracy-focused versus novelty-focused […]

Ver mais

arXiv:2602.18602v1 Announce Type: new Abstract: Package managers are legion. Every programming language and operating system has its own solution, each with subtly different semantics for dependency resolution. This fragmentation prevents multilingual projects from expressing precise dependencies across language ecosystems; it leaves external system and hardware dependencies implicit and unversioned; it obscures security vulnerabilities that lie in the full dependency graph. We present the textit{Package Calculus}, a formalism for dependency resolution that unifies the core semantics of diverse package […]

Ver mais

arXiv:2510.08554v3 Announce Type: replace-cross Abstract: Diffusion language models (DLMs) enable parallel, order-agnostic generation with iterative refinement, offering a flexible alternative to autoregressive large language models (LLMs). However, adapting reinforcement learning (RL) fine-tuning to DLMs remains an open challenge because of the intractable likelihood. Pioneering work such as diffu-GRPO estimated token-level likelihoods via one-step unmasking. While computationally efficient, this approach is severely biased. A more principled foundation lies in sequence-level likelihoods, where the evidence lower bound (ELBO) serves as […]

Ver mais

arXiv:2602.21055v1 Announce Type: cross Abstract: In many applications, weighted networks are constructed based on time series data: each time series is associated to a vertex and edge weights are given by pairwise correlations. The result is a network whose edge dependency structure violates the assumptions of most common network models. Nonetheless, it is common to analyze these “correlation networks” using embedding methods derived from edge-independent network models, based on a belief that the edges are approximately independent. In […]

Ver mais

Can general-purpose AI architectures go beyond prediction to discover the physical laws governing the universe? True intelligence relies on "world models" — causal abstractions that allow an agent to not only predict future states but understand the underlying governing dynamics. While previous "AI Physicist" approaches have successfully recovered such laws, they typically rely on strong, domain-specific priors that effectively "bake in" the physics. Conversely, Vafa et al. recently showed that generic Transformers fail to acquire these world models, […]

Ver mais

Reinforcement learning from human feedback (RLHF) replaces hard-to-specify rewards with pairwise trajectory preferences, yet regret-oriented theory often assumes that preference labels are generated consistently from a single ground-truth objective. In practical RLHF systems, however, feedback is typically emph{multi-source} (annotators, experts, reward models, heuristics) and can exhibit systematic, persistent mismatches due to subjectivity, expertise variation, and annotation/modeling artifacts. We study episodic RL from emph{multi-source imperfect preferences} through a cumulative imperfection budget: for each source, the total deviation of its […]

Ver mais