PARADIGM_SHIFT PARADIGM_SHIFT
329 papers · Page 3 of 4
This paper moves LLMs from point predictions to set-valued predictions with rigorous statistical coverage guarantees.
AI & ML arxiv | Mar 25
Connects stochastic optimal control to the Schrödinger equation, enabling analytic solutions for long-horizon problems that previously scaled exponentially with difficulty.
AI & ML arxiv | Mar 25
Enables 3D medical image segmentation pre-training using only mathematical formulas and implicit functions, requiring zero real-world data or expert annotations.
AI & ML arxiv | Mar 25
A dual-path architecture that combines speculative speech-to-speech prefixes with cascaded LLM continuations for zero-latency, high-quality dialogue.
AI & ML arxiv | Mar 25
A biology-native transformer architecture that mirrors cellular transcription and translation, enabling interpretable predictions across DNA, RNA, and protein.
AI & ML arxiv | Mar 25
Introduces a 'geospatial model foundry' that learns unified representations from the weights of existing models rather than raw data.
AI & ML arxiv | Mar 25
Enables training of monocular novel-view synthesis models using entirely unpaired, in-the-wild internet images.
AI & ML arxiv | Mar 25
Provides a statistically rigorous framework to evaluate model performance and reliability after cherry-picking or selecting models based on the same test data.
AI & ML arxiv | Mar 25
Logical reasoning in LLMs is causally linked to 'algebraic divergence' in the residual stream, and failure to achieve this geometry explains sycophancy.
AI & ML arxiv | Mar 26
Environment Maps nearly double the success rate of long-horizon agents by replacing session-bound context with a persistent, structured graph representation.
AI & ML arxiv | Mar 26
A statistical physics framework that predicts the fundamental limits of agentic self-improvement and nested LLM architectures.
AI & ML arxiv | Mar 26
Bio-inspired visual servoing that achieves low-latency robotic control by processing event-stream flux directly, bypassing traditional state estimation.
AI & ML arxiv | Mar 26
A massive empirical study of 177,000 tools reveals a rapid shift in the AI agent ecosystem from 'perception' to 'action' (27% to 65% usage).
AI & ML arxiv | Mar 26
A simple perturbation method reveals that representations are not just activation patterns, but conduits that determine how learning 'infects' similar examples.
AI & ML arxiv | Mar 26
LLMs can solve planning problems with state spaces as large as 10^165 by acting as program generators rather than direct planners.
AI & ML arxiv | Mar 26
LLM-generated summaries can produce patient embeddings that are more 'portable' and robust to hospital distribution shifts than specialized clinical models.
AI & ML arxiv | Mar 26
Formalizes 'likelihood hacking,' a failure mode where RL-trained models learn to generate unnormalized probabilistic programs to artificially inflate rewards.
AI & ML arxiv | Mar 26
A model-agnostic framework to boost time-series forecasting by aligning internal representations with those of pretrained foundation models.
AI & ML arxiv | Mar 26
Unifies input and predicted meshes under a shared topological framework to enable high-fidelity 3D reconstruction with sharp features.
AI & ML arxiv | Mar 26
Quantifies an emergent 'self' in robots as an invariant subnetwork that persists across continual learning of variable tasks.
AI & ML arxiv | Mar 26
Moves automated research from stateless linear pipelines to a persistent Research World Model that maintains a self-correcting knowledge graph of gaps and methods.
AI & ML arxiv | Mar 26
Introduces a 'sorry-driven' formal decomposition that allows LLM agents to solve complex proofs by isolating and independently verifying subgoals.
AI & ML arxiv | Mar 26
Enforces hard incompressibility constraints in neural operators using spectral Leray projection, ensuring physically admissible fluid simulations.
AI & ML arxiv | Mar 26
LensWalk introduces a 'reason-plan-observe' loop that allows agents to dynamically control the temporal sampling and density of the videos they analyze.
AI & ML arxiv | Mar 26
The Free-Market Algorithm (FMA) is a zero-parameter metaheuristic that discovers complex pathways in chemistry and economics through emergent supply-and-demand dynamics.
AI & ML arxiv | Mar 26
MARCH eliminates 'LLM-as-a-judge' confirmation bias by using information asymmetry to force verification agents to check claims without seeing the original response.
AI & ML arxiv | Mar 26
Shifts AI evaluation from static benchmarks to interactive agentic environments requiring fluid adaptation.
AI & ML arxiv | Mar 27
Provides the first formal proof and verification framework for agent-tool integration protocols.
AI & ML arxiv | Mar 27
Demonstrates that visual hierarchies require Lorentzian causal structure rather than Euclidean space.
AI & ML arxiv | Mar 27
Proves that Transformers can internalize complex search algorithms like MCTS directly into their weights.
AI & ML arxiv | Mar 27
Introduces a multi-answer RL objective that trains models to represent a distribution of valid answers in a single forward pass.
AI & ML arxiv | Mar 27
The 'Reasoning Contamination Effect' shows that Chain-of-Thought (CoT) reasoning actually disrupts a model's internal confidence signal, leading to poorer calibration.
AI & ML arxiv | Mar 27
R1Sim applies the 'Reasoning-RL' paradigm (popularized by DeepSeek-R1) to traffic simulation, achieving superior safety and diversity in multi-agent behaviors.
AI & ML arxiv | Mar 27
SIGMA resolves 'trajectory divergence' in molecular string generation by enforcing geometric symmetry recognition through contrastive learning.
AI & ML arxiv | Mar 27
Pixelis shifts VLM reasoning from static description to a 'reasoning in pixels' agentic paradigm that learns via an executable tool grammar.
AI & ML arxiv | Mar 27
The AE4E paradigm proposes a 'Social Contract' for multi-agent economies, replacing individual model alignment with an institutional 'Separation of Power'.
AI & ML arxiv | Mar 27
Using Signal Detection Theory, this work proves that LLM calibration and 'metacognitive efficiency' (knowing what you know) are distinct, dissociable capacities.
AI & ML arxiv | Mar 27
Vision Hopfield Memory Networks (V-HMN) present a brain-inspired alternative to Transformers and Mamba using hierarchical associative memory mechanisms.
AI & ML arxiv | Mar 27
Representing GPS trajectories as hyperspectral images enables multi-month dense anomaly detection that was previously computationally intractable.
AI & ML arxiv | Mar 27
Fixes the inherent instability of on-policy distillation in LLMs using local support matching and top-p rollout sampling.
AI & ML arxiv | Mar 27
Enables LMMs to 'think' using compact latent visual representations rather than verbalizing everything into text.
AI & ML arxiv | Mar 27
Introduces the concept of a 'trainable' knowledge base for RAG that improves performance by distilling and writing back compact knowledge units.
AI & ML arxiv | Mar 27
Uses cycle-consistency as a label-free reward signal for reinforcement learning to resolve contradictions in multimodal reasoning.
AI & ML arxiv | Mar 27
Introduces a CNN architecture where feature maps are mathematically identical to Grad-CAM saliency maps by design, rather than post-hoc.
AI & ML arxiv | Mar 30
Shifts world model evaluation from visual fidelity to 'Simulative Reasoning,' revealing a massive gap in current AI's ability to plan.
AI & ML arxiv | Mar 30
Learns high-level symbolic state machines directly from raw pixels to guide robot control without hand-crafted priors.
AI & ML arxiv | Mar 30
Demonstrates that symbolic event primitives (like Schank's Conceptual Dependency) can be 'rediscovered' by neural networks purely through compression pressure.
AI & ML arxiv | Mar 30
Identifies specific hidden-state dimensions (H-Nodes) responsible for hallucinations and introduces a real-time defense to cancel them.
AI & ML arxiv | Mar 30
Moves industrial recommendation systems from static multi-stage pipelines to self-evolving agentic loops.
AI & ML arxiv | Mar 30
Empirically proves that AI Scientist agents can genuinely learn from physical experimental feedback via in-context learning.
AI & ML arxiv | Mar 30
Replaces standard autoregressive action generation in robot VLAs with iterative refinement via discrete flow matching.
AI & ML arxiv | Mar 30
Introduces a multi-agent CAD generation pipeline that uses programmatic geometric validation from the OpenCASCADE kernel to iteratively fix dimensional errors.
AI & ML arxiv | Mar 30
Introduces Process-Aware Policy Optimization (PAPO) to solve the chronic issue of reward hacking in process reward models (PRMs).
AI & ML arxiv | Mar 30
Demonstrates that perplexity/log-likelihood is a deceptive metric for model distillation, often masking massive drops in actual generation quality.
AI & ML arxiv | Mar 30
Shifts 3D scene generation from diffusion to a fully autoregressive paradigm using next-token prediction of 3D Gaussian primitives.
AI & ML arxiv | Mar 30
Proposes a universal denoiser that outperforms the Bayes-optimal Tweedie's formula when the noise distribution is unknown.
AI & ML arxiv | Mar 30
Introduces geometry-aware parallel refinement for diffusion language models, bypassing fixed-block decoding limitations.
AI & ML arxiv | Mar 31
Knowledge distillation can be performed by injecting 'experience' into prompts rather than updating model weights.
AI & ML arxiv | Mar 31
Gaussian Joint Embeddings provide a probabilistic alternative to deterministic SSL, eliminating the need for architectural asymmetries to prevent collapse.
AI & ML arxiv | Mar 31
Identifies a 'stability asymmetry' signature where deceptive models maintain stable internal beliefs while producing fragile, unstable external responses under perturbation.
AI & ML arxiv | Mar 31
Challenges the 'filter-first' data paradigm by showing that training on uncurated data with quality-score labels outperforms training on high-quality filtered subsets.
AI & ML arxiv | Mar 31
Introduces a 'clone-robust' mechanism (YRWR) to prevent AI model producers from strategically gaming the rankings in crowd-sourced arenas like Chatbot Arena.
AI & ML arxiv | Mar 31
Introduces neural topology probing to identify causally influential 'hub neurons' in Vision-Language Models that govern cross-modal behavior.
AI & ML arxiv | Mar 31
Proposes a new reinforcement learning policy compression method based on long-horizon state-space coverage instead of immediate action-matching.
AI & ML arxiv | Mar 31
Identifies that standard Transformer attention matrices are fundamentally ill-conditioned and proposes a drop-in 'preconditioned' replacement.
AI & ML arxiv | Mar 31
Challenges the necessity of discrete action tokenizers in robotics by using a continuous, single-stage flow matching policy.
AI & ML arxiv | Mar 31
Introduces a marketplace infrastructure that rebrands AI agents from mere tools into peer participants in a verifiable production network.
AI & ML arxiv | Mar 31
Introduces a vision model testbed that aligns AI visual attention (scanpaths) with human gaze without sacrificing classification accuracy.
AI & ML arxiv | Mar 31
Collapses the standard vision backbone-plus-decoder architecture into a single early-fusion Transformer stack for both perception and task modeling.
AI & ML arxiv | Mar 31
Couples visual representations directly into the RL optimization process (RLVR) for vision-language models using a structured reward reweighting mechanism.
AI & ML arxiv | Mar 31
Proposes 'Amdahl’s Law for AI,' proving that human effort in AI-assisted work is bottlenecked by the fraction of 'novel' tasks rather than agent capability.
AI & ML arxiv | Mar 31
Shifts protein fitness optimization from continuous embeddings to discrete Quadratic Unconstrained Binary Optimization (QUBO).
AI & ML arxiv | Mar 31
Introduces LongCat-Next, a 'Native Multimodal' model that treats vision and audio as first-class discrete tokens rather than language-centric attachments.
AI & ML arxiv | Mar 31
Proposes SOL-Nav, which replaces raw visual features in navigation with structured language descriptions for LLM-based agents.
AI & ML arxiv | Mar 31
Sci-Mind introduces an 'Adversarial Cognitive Dialectic' where specialized agents debate to refine mathematical models.
AI & ML arxiv | Mar 31
Introduces 'Umwelt Engineering,' the deliberate constraint of an agent's linguistic environment to improve reasoning.
AI & ML arxiv | Mar 31
Introduces Composer, a paradigm that generates input-specific parameter adaptations at inference time to enable dynamic per-input model specialization.
AI & ML arxiv | Mar 31
SkyNet extends MuZero to partially-observable stochastic games by adding auxiliary belief-aware heads, significantly outperforming baselines in complex card games.
AI & ML arxiv | Mar 31
The Physics-Guided Transformer (PGT) embeds physical priors (like diffusion and causality) directly into the self-attention mechanism via heat-kernel biases.
AI & ML arxiv | Mar 31
SARL improves reasoning models by rewarding the 'topology' of thoughts rather than just the final answer, enabling effective RL without ground-truth labels.
AI & ML arxiv | Mar 31
Correlated Diffusion replaces independent noise with structured MCMC dynamics, enabling generative modeling on hyper-efficient probabilistic computers.
AI & ML arxiv | Mar 31
This paper clarifies that Diffusion Maps (DMAPs) are not actually a dimensionality reduction tool, but rather a spectral representation that requires specific combinations to form a chart.
AI & ML arxiv | Mar 31
PhysNet embeds physical tumor growth dynamics directly into the latent feature space of a CNN, rather than just as a constraint on the output.
AI & ML arxiv | Mar 31
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.
AI & ML arxiv | Mar 31
Moves VLM grounding from text-based coordinates to a direct visual token selection mechanism via special pointing tokens.
AI & ML arxiv | Mar 31
Bypasses expensive formal verification solvers by designing neural networks that are 'verifiable by design' using the fast trivial Lipschitz bound.
AI & ML arxiv | Mar 31
Replaces traditional fixed-update rules in online learning with a causal Transformer to track switching experts in non-stationary environments.
AI & ML arxiv | Mar 31
Moves beyond next-token prediction to model reasoning as gradient-based energy minimization over latent trajectories.
AI & ML arxiv | Mar 31
Entropic Claim Resolution (ECR) shifts RAG from retrieving 'relevant' documents to retrieving 'discriminative' evidence that minimizes hypothesis uncertainty.
AI & ML arxiv | Mar 31
The 'Bidirectional Coherence Paradox' demonstrates that LLM performance and explanation quality can be inversely correlated depending on domain observability.
AI & ML arxiv | Mar 31
COvolve creates an automated curriculum for open-ended learning by co-evolving environments and policies as executable code through a zero-sum game.
AI & ML arxiv | Mar 31
Seen2Scene is the first flow matching model trained directly on incomplete real-world 3D scans rather than synthetic complete data.
AI & ML arxiv | Mar 31
Unrestrained Simplex Denoising treats discrete data generation as a non-Markovian process on the probability simplex.
AI & ML arxiv | Mar 31
PRCO decouples perception and reasoning in Multimodal RL through an Observer-Solver architecture.
AI & ML arxiv | Mar 31
SOLE-R1 uses Vision-Language Model chain-of-thought reasoning as the sole reward signal for zero-shot robotic reinforcement learning.
AI & ML arxiv | Mar 31
Metric Similarity Analysis (MSA) uses Riemannian geometry to compare the intrinsic geometry of neural representations.
AI & ML arxiv | Mar 31
Replaces the heuristic constant momentum (0.9) with a parameter-free, physics-inspired schedule that speeds up convergence by nearly 2x.
AI & ML arxiv | Apr 1
Proposes a mathematical framework where 'spectral gaps' in parameter updates control phase transitions like grokking and loss plateaus.
AI & ML arxiv | Apr 1
Proposes a neuroscience-grounded memory architecture that makes interactions cheaper and more accurate with experience, rather than relying on expanding context windows.
AI & ML arxiv | Apr 1
Introduces DASES, a framework that replaces passive validation with active 'falsification' to ensure scientific models learn actual mechanisms rather than just winning benchmarks.
AI & ML arxiv | Apr 1