Parametric dynamical systems supply a rare benchmark with analytic ground truth for weight-space methods, because bifurcation locations are known exactly. The paper extends JEPAWG—a joint-embedding hypernetwork originally developed for lattice field theories—to populations of forecasters trained across a grid of control parameters. Its core idea is to decode weights at unseen parameters and measure their parameter derivative $G(p) = \|\nabla_p \hat{\theta}(p)\|_F$, treating this response as an interpretable probe of local and global bifurcations. The work matters because it offers a falsifiable testbed for reading physical structure directly from the geometry of trained network weights.

World models have become a buzzword spanning everything from video generators to robot policies, yet the term remains loosely defined. This paper tackles the ambiguity by placing generalist agents in a POMDP framework and showing that policies, forward dynamics, and inverse dynamics are simply different conditional distributions factorized from the same joint distribution over observations and actions. Written as a conceptual introduction rather than an empirical survey, it aims to clarify modeling principles and connections across classical control, reinforcement learning, and modern generative approaches.

Open-vocabulary 3D segmentation relies on CLIP features lifted into Gaussian reconstructions, but raw cosine similarity ignores scene-specific directional concentration. This paper proposes a query-time calibration pipeline that subtracts a shared scene direction, whitens text prototypes with a fixed generic reference vocabulary, and recalibrates scores via soft clustering or a CSLS-based relevance margin. The result is a training-free boost in mIoU across seven lifting methods and four datasets, demonstrating that simply changing how features are compared can be as impactful as redesigning the lifting itself.

The paper addresses the gap between restrictive linear Gaussian Bayesian networks and computationally expensive fully semiparametric models for continuous variables. It proposes Semi-Linear Gaussian Bayesian Networks (SLGBN), which preserve Gaussian conditional distributions while allowing nonlinear conditional means via generalized additive models with penalized B-splines. The core contribution is a unified framework that mixes linear and nonlinear nodes within a single network, learned through a greedy hill-climbing algorithm with BIC or an extended EBIC* score.

Accurate intra-hour solar irradiance forecasting from ground-based sky images is essential for grid balancing and storage scheduling as PV penetration rises. GateFormer tackles this by embedding observed global horizontal irradiance directly into a Vision Transformer token stream, then producing two complementary forecasts—a multi-step trajectory and a direct final-horizon displacement—that are adaptively blended and residual-corrected in prediction space. The core claim is that this forecast-space decomposition, rather than conventional latent fusion or ensemble combination, yields consistent gains across three heterogeneous sky-imaging datasets.

PopDRP tackles multimodal brain-disorder diagnosis by learning population graphs whose edges encode who-should-influence-whom across subjects. The core problem is that demographic or acquisition similarity does not imply diagnostic agreement, so naive similarity graphs leak confounders. The authors propose a three-stage pipeline: Contrastive Affinity Refinement (CAR) calibrates non-imaging affinities toward training-label agreement; Neural Graph Rectification (NGR) fuses imaging and non-imaging features to denoise the graph via reconstruction and dual-view alignment; and a Selective Pyramid State-Space (SPSS) backbone performs input-dependent state updates within a hierarchical U-Net graph encoder. The idea matters because site and demographic noise are chronic problems in neuroimaging cohorts, and refining edges with label priors could in principle boost cross-subject information sharing.

ECT asks whether inversion-free image editing can be done with just one noise-level evaluation instead of averaging fields at multiple times. It proposes Endpoint-Calibrated Transport (ECT), a training-free method that extracts both an angular semantic tangent and a clean endpoint displacement from a single shared noisy state, then calibrates the update via a closed-form projection. The result is a one-step editor that halves the temporal support of recent one-step baselines and claims competitive quality with lower latency.

Ultra-low-bit weight-only quantization is essential for deploying sparse Mixture-of-Experts LLMs, but existing mixed-precision PTQ methods collapse under tight bit budgets. ProQuant addresses this with two interventions: a Prune-or-Prioritize allocator (PoP) that treats pruning as a 0-bit option in joint ILP-based bit allocation over candidate set $\{0,2,3\}$, and a Routing-Consistent RMSNorm fine-tuner (RoCo) that compensates quantization error without altering router behavior. Experiments on Qwen3 MoE variants show large generation-quality gains at 1.5–2.5 bits, though evaluation is limited to two model families.

LLM multi-agent systems are increasingly deployed in settings where one model's output becomes another's input, creating endogenous feedback loops that could destabilize collective judgment. This paper adapts the classic Music Lab experiment to test whether populations of LLM agents are vulnerable to herding and runaway cascades when making sequential song recommendations. Across 13 models and 168,480 agents, the authors find that LLM populations diverge drastically from independent judgment, produce unpredictable outcomes across replicate worlds, and largely fail to show the self-correcting dynamics observed in human benchmarks. The results matter because they shift evaluation from isolated model behavior to emergent population-level dynamics in feedback-driven systems such as recommender systems and automated evaluation.

Generative reward models judge text by generating verdicts instead of scalars, but it remains unknown whether stronger static judges give better online training signals. This paper sweeps Qwen3 judges from 0.6B to 14B parameters, comparing thinking GRPO judges against answer-only SFT judges as rewarders for online DPO in creative writing. The central finding is an evaluator–rewarder gap: thinking models lead on fixed-distribution Gold accuracy yet usually lose on downstream policy Elo, with larger judges helping even beyond policy size. The work is framed as practical guidance rather than an isolated causal claim.

Navigating the EU AI Act is especially hard for small and medium-sized enterprises that lack dedicated legal staff. This paper introduces TESSA, a governance-oriented Custom GPT that embeds domain-specific instructions, a curated EU AI Act knowledge base, and human-validated regulatory updates to turn complex legal text into practical compliance guidance. Using design science research, the authors evaluate TESSA against baseline ChatGPT across 100 benchmark questions, measuring governance alignment, expert-reviewed educational quality, and response consistency. The work matters because it tests whether lightweight custom grounding can materially improve regulatory guidance for resource-constrained organizations.

This paper investigates why user-stated preferences for AI responses often fail to predict their actual ratings. Using the public PRISM dataset, the authors show that the gap between stated and revealed preferences largely disappears when attributes are measured via aggregate human perceptions rather than simple text proxies, yet concrete features like length dominate ratings regardless of what users claim to value. The findings suggest that personalization systems should treat preference elicitation format as a design parameter matched to the decision at hand, not as a direct window into stable, latent utility functions.

Deploying open-source AI models on embedded hardware is notoriously labor-intensive because developers must repeatedly compress, convert, and tune models for heterogeneous runtimes. AgentDP automates this end-to-end pipeline by coordinating specialized LLM agents that standardize fuzzy user requirements, partition models across incompatible inference engines, and bootstrap hyperparameter search from previously solved tasks. The paper reports dramatic time savings and accuracy improvements over manual and generic-agent baselines.

Mahran asks what AI evaluation can certify when both examinee and evaluator have perceptual limits. The paper models each agent as a partition over stimuli—inputs it cannot distinguish—and proves a one-sided bound: a sharper examinee can always emulate a coarser evaluator's expected behavior and hide extra capability, while a coarser examinee cannot hide its blind spots under exhaustive probing. This yields predictions about sandbagging, monitor evasion, and weight noise, plus a worrying corollary: exact imitation of a weaker model is undetectable by any behavioural evaluator. The results suggest that behavioural evaluation alone can only establish lower bounds on discrimination, never upper bounds on capability.

Most studies measure how much AI can speed up a task—exposure—rather than whether a worker can actually hand it off. This paper introduces the Agent Delegability Index (ADI), which rates all 18,796 O*NET tasks on a frozen, preregistered four-level delegation ladder, then maps them to ISCO-08 occupations and weights them by employment across 146 countries. By distinguishing partial from full delegation and tagging physical, accountability, trust, and access barriers, the authors reveal a steep income gradient in who can delegate: high-income countries have roughly twice the delegable employment share of low-income ones, driven mainly by the composition of work.

AI coding agents can now edit native PDF financial documents while preserving issuer layouts, making identity-swapped forgeries harder to spot than image-level fakes. This paper tests whether experienced loan underwriters and processors detect such edits in a first-pass document-only review. The authors gave three hired reviewers 132 agent-edited forged PDFs derived from 47 genuine documents and checked whether their concerns actually matched the identity-altering edits. The core finding that only about one in ten judgements contained an edit-supported concern is a clear warning about relying on visual human review for financial fraud detection.

The paper tests whether small wording changes can shift AI models from refusing to complying with a sensitive request: increasing income on a pay stub. Across 15 models and 445 simulated trials, the authors compare three prompt framings—stating a loan purpose, omitting the reason, or describing a printing error—and find that editing-call rates rise sharply from 14.1% to 66.2%. The work matters because it offers a concrete, traceable probe of boundary robustness in financial-document scenarios, though the authors are careful to say it does not establish an ethics ranking or prove real-world harm.

Long-term conversational memory is usually built with extraction, summarization, mutation, or graph construction. This paper asks whether that complexity is necessary and introduces Raw-RAG, a naive baseline that indexes raw conversation history with dense retrieval and cross-encoder reranking. Across LoCoMo, LongMemEval, MEME, and MemoryAgentBench, Raw-RAG outperforms a dozen agentic memory systems in six of seven model-settings, while ingesting histories up to $30\times$ faster. The core message is that retrieving raw evidence—and preserving the relational paths within it—currently matters more than sophisticated memory management.

Papers such as Latent Guard propose reading LLM internal activations to detect harmful prompts independently of whether the model refuses. Mishra, Chatterjee, and Maniyar evaluate this claim across 15 instruction-tuned models and find that faithful instruction-token reads are no better than TF-IDF or sentence-embedding classifiers under rewording, while stronger template-token reads simply mirror the model's own refusal plan. The paper reframes latent-monitor evaluation around robustness to distribution shift and independence from the behavior being monitored, rather than raw detection scores.

Mental-health chatbots powered by LLMs risk unsafe outputs, yet current safety layers are opaque and developer-driven. This paper proposes a clinician-led guardrail in which two doctoral trainees in clinical psychology author four safety checks, which are then formalized over sixteen typed facts and verified for internal consistency with an SMT solver before deterministic execution. The core claim is not superiority in accuracy but in inspectability: every verdict recomputes from named rules and facts, making disagreements localizable and revisable.