Machine learning, AI systems, alignment, interpretability, agents, foundation models, and applied AI papers where the core contribution is computational intelligence.
Filter by category: Paradigm Challenge Breaks Assumption First Ever Nature Is Weird Practical Magic Cosmic Scale Life Origin Open Release Efficiency Leap New Capability Scaling Insight
New Capability
The first framework for bit-identical deep learning training that produces MD5-verified identical weights across independent runs.
Efficiency Breakthrough
Drift-AR enables single-step (1-NFE) high-fidelity image generation by reinterpreting AR prediction entropy as a physical drifting field.
New Capability
Meta-Harness automates the engineering of the 'code' surrounding LLMs, improving RAG and agent performance by optimizing retrieval and context management logic.
Efficiency Breakthrough
ROVED reduces the expensive human feedback required for preference-based RL by up to 90% by leveraging vision-language embeddings and uncertainty filtering.
Paradigm Shift
PhysNet embeds physical tumor growth dynamics directly into the latent feature space of a CNN, rather than just as a constraint on the output.
Paradigm Shift
This paper proves that reward hacking is a structural equilibrium of optimized AI agents, not a bug, and provides a computable 'distortion index' to predict it.
Paradigm Shift
Moves VLM grounding from text-based coordinates to a direct visual token selection mechanism via special pointing tokens.
Efficiency Breakthrough
Introduces Heddle, a trajectory-centric system that resolves the long-tail latency bottleneck of tool calls in agentic Reinforcement Learning.
Paradigm Shift
Bypasses expensive formal verification solvers by designing neural networks that are 'verifiable by design' using the fast trivial Lipschitz bound.
New Capability
A training-free metacognitive framework that gives LLMs explicit control over expanding, pruning, and repairing reasoning trajectories during inference.
New Capability
Presents PReD, the first foundation model and 1.3M-sample dataset specifically for electromagnetic signal perception and decision-making.
Paradigm Shift
Replaces traditional fixed-update rules in online learning with a causal Transformer to track switching experts in non-stationary environments.
Efficiency Breakthrough
Replaces the classic Newton-Raphson power-flow solver with a differentiable GPU-accelerated simulation.
New Capability
Transitions reasoning model optimization from coarse sequence-level advantages to fine-grained token dynamics.
Paradigm Shift
Moves beyond next-token prediction to model reasoning as gradient-based energy minimization over latent trajectories.
Efficiency Breakthrough
Introduces lightweight equilibration to the Muon optimizer, significantly stabilizing and accelerating LLM pretraining.
Scaling Insight
Discovers that LLM hidden states undergo geometric 'warping' at digit-count boundaries, mimicking human psychological perception.
Efficiency Breakthrough
Enables instruction-following in low-resource languages by simply merging target language base models with English-instructed models.
New Capability
Enhances Kolmogorov-Arnold Networks (KAN) with fractal interpolation to approximate non-smooth and rough functions.
Breaks Assumption
Exposes a massive robustness gap in Vision-Language-Action (VLA) models, where simple paraphrasing causes up to 50% success drops.
Efficiency Breakthrough
An evolutionary framework for GPU kernel generation that outperforms frontier models like Claude 4.6 and Gemini 3.0.
Efficiency Breakthrough
HISA eliminates the quadratic O(L²) bottleneck in sparse attention indexers, enabling efficient long-context scaling for models like DeepSeek-V3.
New Capability
Researchers have used LLMs to evolve entirely new Reinforcement Learning update rules from scratch that compete with human-designed baselines like PPO and SAC.
Breaks Assumption
The 'Scaffold Effect' reveals that Vision-Language Models in clinical settings often fabricate reasoning based on prompt framing rather than actual visual data.
Paradigm Shift
Entropic Claim Resolution (ECR) shifts RAG from retrieving 'relevant' documents to retrieving 'discriminative' evidence that minimizes hypothesis uncertainty.
Efficiency Breakthrough
IsoQuant leverages SO(4) isoclinic rotations to achieve a 4.5x-4.7x speedup in low-bit KV-cache quantization over existing methods.
Paradigm Shift
The 'Bidirectional Coherence Paradox' demonstrates that LLM performance and explanation quality can be inversely correlated depending on domain observability.
Paradigm Shift
COvolve creates an automated curriculum for open-ended learning by co-evolving environments and policies as executable code through a zero-sum game.
Efficiency Breakthrough
INSID3 achieves state-of-the-art one-shot image segmentation using only frozen DINOv3 features without any training, fine-tuning, or auxiliary models.
Efficiency Breakthrough
EdgeDiT provides a hardware-aware blueprint for running massive Diffusion Transformers (DiT) on mobile NPUs with a 1.6x reduction in latency.
Efficiency Breakthrough
LAD achieves 3x lower latency than previous driving language models by generating textual reasoning and motion plans at up to 20 Hz.
New Capability
The TAG glove system provides high-resolution tactile feedback and precise 21-DoF motion capture for under $1000.
Paradigm Shift
Seen2Scene is the first flow matching model trained directly on incomplete real-world 3D scans rather than synthetic complete data.
Efficiency Breakthrough
Hydra unifies ColBERT-style retrieval and autoregressive generation into a single Vision-Language Model using a single LoRA adapter.
Efficiency Breakthrough
StreamingVLA eliminates execution halting in robots by asynchronously parallelizing observation, generation, and execution.
Paradigm Shift
Unrestrained Simplex Denoising treats discrete data generation as a non-Markovian process on the probability simplex.
New Capability
SPINNER is a tri-rotor UAV that uses continuous self-rotation to expand the field of view of its sensors without adding extra hardware.
New Capability
Medical AI Scientist is the first autonomous framework for clinically grounded research ideation and manuscript drafting.
Efficiency Breakthrough
ResAdapt learns a per-frame visual budget allocator that optimizes input resolution before encoding.
Breaks Assumption
LACE enables continual learning models to automatically expand their own capacity by monitoring loss signals during training.
Paradigm Shift
PRCO decouples perception and reasoning in Multimodal RL through an Observer-Solver architecture.
Scaling Insight
This paper establishes the formal information-theoretic limits and conditions under which self-improving AI systems can be safely verified.
Efficiency Breakthrough
RNNs can be trained online without Jacobian propagation, matching BPTT performance at 1000x less memory.
Breaks Assumption
Sparse Autoencoders (SAEs) fail at compositional generalization due to flawed dictionary learning, not the inference method.
Paradigm Shift
SOLE-R1 uses Vision-Language Model chain-of-thought reasoning as the sole reward signal for zero-shot robotic reinforcement learning.
Efficiency Breakthrough
IF4 introduces an adaptive 4-bit data type that switches between Float and Integer representations to minimize quantization error.
New Capability
Vision-Language Models (VLMs) can outperform specialized learning-based placers in chip floorplanning through visual evolutionary optimization.
Scaling Insight
HyperP provides the first hyperparameter transfer laws for hypersphere optimization, ensuring stable scaling for models using the Muon optimizer.
New Capability
DreamLite enables sub-second 1024x1024 image generation and editing on mobile devices using a unified 0.39B parameter model.
Paradigm Shift
Metric Similarity Analysis (MSA) uses Riemannian geometry to compare the intrinsic geometry of neural representations.