AI & Machine Learning

Introduces a framework for LLMs to self-improve reasoning in specific domains by autonomously mining and constructing training environments directly from the open web.

Establishes a formal mathematical equivalence between Classifier-Free Guidance (CFG) and alignment-based objectives, allowing for CFG-like quality without inference-time overhead.

Proposes an agentic architecture that achieves O(1) token complexity relative to dataset size by strictly separating intent parsing from deterministic data execution.

Achieves high-fidelity diffusion generation in just 3 steps by distilling layer-wise time embeddings from reference trajectories.

Finds that nominal instruction-tuning with LoRA often fails to improve (and can even degrade) verifiable instruction-following despite improvements on broader benchmarks.

Shifts symbolic regression from discrete genetic search to a continuous, embedding-driven optimization paradigm.

Reveals that RLVR-driven reasoning improvements in LLMs are the result of highly sparse changes to a tiny fraction of 'critical' token distributions.

Identifies that the full source code (skill body) of a tool is the primary signal for LLM tool selection, far outweighing the importance of descriptions or metadata.

Replaces standard autoregressive document OCR with a parallel diffusion-based denoising framework.

Introduces a verifier that operates directly on the latent hidden states of Diffusion Transformers, avoiding the need for costly pixel-space decoding during inference-time scaling.

Demonstrates that Hebbian plasticity can induce emergent attractor dynamics in robot controllers, enabling rapid adaptation without backpropagation.

Uncovers that neural operator digital twins are acutely vulnerable to sparse adversarial perturbations on boundary conditions that bypass standard anomaly detection.

Leverages unstructured clinical notes during training to boost the performance of models that are deployed using only structured EHR data.

Robotic bipedal mass scales with the square of leg length rather than the cubic scaling found in biological systems.

CanViT is the first task-agnostic active-vision foundation model that reconstructs scenes using low-resolution 'glimpses' with 19.5x fewer FLOPs than existing models.

A large-scale study of 12 reasoning models reveals that internal 'thinking' processes frequently recognize deceptive hints while the final output remains sycophantic.

Instead of using top-activating examples, this method steers Sparse Autoencoder (SAE) features in Vision-Language Models to let the model describe its own internal visual features.

DeIllusionLLM introduces task-level autoregressive reasoning to prevent LLMs from hallucinating answers to ill-posed or faulty scientific questions.

CAM3R is a camera-agnostic 3D reconstruction model that handles fisheye, panoramic, and pinhole imagery without requiring prior calibration.

Inter-Layer Structural Encoders (ILSE) use Cayley graphs to aggregate features from all internal LLM layers, improving accuracy by up to 44% over final-layer-only predictions.

Introduces the first high-performing open-source metric for per-sample AI music quality evaluation.

Provides a massive 2.5M image-to-TikZ dataset and the first instruction-augmented dataset for geometric visual reasoning.

A new statistical test that reliably detects whether a dataset was NOT used in an LLM's training corpus.

Introduces Dual Q-DM, the first non-adversarial imitation learning method theoretically guaranteed to eliminate compounding errors.

A quantitative model that predicts the performance gain of merging independent LLM specialists before committing compute.

Proves that logic and lookup-table (LUT) based neural networks are structurally more resilient to hardware bit-flips than standard architectures.

Identifies the 'Caterpillar Tree' as the theoretically optimal structure for test-time computation and backtracking in LLMs.

ABSTRAL automates the design of multi-agent systems by treating architectures as evolving, inspectable natural-language documents.

Frontier models' reasoning steps are largely 'decorative' and do not causally determine the final answer in most tasks.

Moving beyond coarse reward signals, this paper introduces token-level policy optimization for multimodal reasoning.

UniQueR reconstructs full 3D scenes (including occluded areas) from unposed images in a single forward pass.

Persistent structural memory in neural networks is fundamentally limited by the instability of jointly-learned coordinate systems.

Deep semi-parametric models allow for the instant deletion of training data from a model without retraining or parameter updates.

A 0.26M parameter model using continuous dynamics outperforms 27M parameter recursive models on complex logic tasks like Sudoku-Extreme.

Standard confidence calibration is structurally biased when ground truth labels are ambiguous or annotators disagree.

Agile-VLA enables high-frequency robot control on edge devices by decoupling perception from action through implicit affordance anchoring.

EchoKV introduces a reversible KV cache compression scheme that allows LLMs to switch back to full-precision inference on-demand.

ForestPrune achieves up to 90% token reduction in video MLLMs with minimal accuracy loss using a training-free spatial-temporal forest modeling approach.

Theoretical analysis reveals that the efficiency benefits of low-dimensional data structures for diffusion models diminish significantly when the data manifold is non-linear.

This paper moves LLMs from point predictions to set-valued predictions with rigorous statistical coverage guarantees.

WorldMesh generates consistent, large-scale 3D worlds by populating a geometric mesh scaffold with image diffusion-derived content.

Graph Foundation Models (GFMs) are shown to fail when using fixed architectural backbones, requiring a new approach of inference-time architecture adaptivity.

Access to conversational memory allows an 8B model to outperform a 235B model on user-specific queries while reducing inference costs by 96%.

A rigorous evaluation shows that simple Probabilistic Circuits often outperform complex diffusion-based models for tabular data generation at a fraction of the cost.

Optimizing autoregressive image models with Group Relative Policy Optimization (GRPO) achieves competitive quality without the 2x inference cost of Classifier-Free Guidance.

Identifies that MLLMs fail to perceive visual illusions due to a high-frequency attention bias and provides a plug-and-play fix that boosts accuracy from 13% to 84%.

Polaris introduces a 'Gödel Agent' framework that allows 7B-parameter models to recursively improve their own policies through auditable code patches.

DILLO enables 14x faster safety-critical agent steering by predicting action consequences from latent states instead of heavy visual simulations.

Exposes a major flaw in medical super-resolution research where models trained on downsampled data fail to recover actual lost structures in real low-resolution scans.

Connects stochastic optimal control to the Schrödinger equation, enabling analytic solutions for long-horizon problems that previously scaled exponentially with difficulty.