PARADIGM_SHIFT PARADIGM_SHIFT
329 papers · Page 1 of 4
This paper introduces a graph tokenization framework that allows standard Transformers like BERT to beat specialized Graph Neural Networks without any architectural changes.
AI & ML arxiv | Mar 13
Continual Representation Learning (CoRe) moves PEFT from weight-level updates to representation-space interventions, solving catastrophic forgetting in dynamic environments.
AI & ML arxiv | Mar 13
Theoretical analysis proves that Langevin dynamics is fundamentally non-robust to score function errors, justifying the shift to Diffusion Models.
AI & ML arxiv | Mar 13
HAPO resolves the advantage collapse problem in sparse-reward RL for reasoning models using a Thompson-sampled hindsight mechanism.
AI & ML arxiv | Mar 13
This paper introduces Finsler geometry to manifold learning, allowing for the capture of asymmetric data relationships like density hierarchies that Riemannian methods ignore.
AI & ML arxiv | Mar 13
Manifold-Optimal Guidance reformulates Classifier-Free Guidance (CFG) as a Riemannian control problem, eliminating the artifacts and saturation typical of high guidance scales.
AI & ML arxiv | Mar 13
Expert Threshold Routing (ET) replaces standard top-k token-choice with an independent thresholding mechanism, achieving 1.6x faster training convergence.
AI & ML arxiv | Mar 13
Introduces the Compression-Consistency Principle, arguing that LLMs prefer truth only when false alternatives are structurally harder to compress.
AI & ML arxiv | Mar 13
Enables agents to autonomously discover the group structure of their environments to learn disentangled representations without human priors.
AI & ML arxiv | Mar 13
Eliminates lookahead bias in financial backtesting through a series of yearly-partitioned pretrained LLMs.
AI & ML arxiv | Mar 13
Solves GNN over-squashing by using global effective resistance to identify and rewire structural bottlenecks.
AI & ML arxiv | Mar 13
Proposes a unified image tokenizer that reconciles the conflicting requirements of visual understanding and generation using a residual evolution process.
AI & ML arxiv | Mar 13
Introduces a feature-matching objective for LLM fine-tuning that targets sequence-level statistics without requiring reward models or ground-truth verifiers.
AI & ML arxiv | Mar 13
Demonstrates that the stochasticity in standard regularized model training (like cross-validation) can serve as a 'free' and effective exploration strategy for contextual bandits.
AI & ML arxiv | Mar 13
This paper proposes a method to align and personalize LLMs directly from raw user interactions using self-distillation, bypassing the need for explicit human labels or RLHF.
AI & ML arxiv | Mar 16
Introduces the Budget-Sensitive Discovery Score (BSDS), a formally verified metric machine-checked in Lean 4 for evaluating AI-guided scientific candidate selection.
AI & ML arxiv | Mar 16
This paper establishes a systematic protocol for 'stitching' heterogeneous Vision Foundation Models (e.g., CLIP and DINOv2) to share early layers while retaining specialized capabilities.
AI & ML arxiv | Mar 16
Introduces Modal Logical Neural Networks (MLNNs) as a differentiable logic layer that bridges deep learning with symbolic Kripke semantics for regulated AI.
AI & ML arxiv | Mar 16
Demonstrates a robot that improves its own locomotion by identifying and physically 'self-destructing' redundant or inhibiting limbs during its lifetime.
AI & ML arxiv | Mar 16
Derives an exact, unbiased policy gradient for Reinforcement Learning on Diffusion LLMs, bypassing the need for sequence-level likelihood approximations.
AI & ML arxiv | Mar 16
Proposes modeling the world in the feature space of frozen geometry foundation models instead of pixels, achieving 5x faster depth forecasting.
AI & ML arxiv | Mar 16
A small-scale molecular reasoning model that outperforms ultra-large foundation models via structured chain-of-thought and RL.
AI & ML arxiv | Mar 16
ThinkStream introduces a 'Watch-Think-Speak' paradigm for video reasoning that allows models to incrementally update understanding and decide when to respond in real-time.
AI & ML arxiv | Mar 16
Connects DDIM reverse chains to fractal geometry, providing a mathematical explanation for why diffusion models switch from global context to local detail.
AI & ML arxiv | Mar 16
Proposes Causal Process Reward (CPR) to fix 'cherry-picking' in MLLM reasoning by coupling answer correctness with step-level logical alignment.
AI & ML arxiv | Mar 16
Reimagines 3D molecules as continuous vector fields rather than discrete graphs, decoupling structure learning from atom types.
AI & ML arxiv | Mar 16
Diffusion LLMs can match autoregressive (AR) reasoning performance by using AR-generated plans as globally visible scaffolds.
AI & ML arxiv | Mar 17
The Spherical Kernel Operator (SKO) replaces dot-product attention with ultraspherical polynomials to bypass the saturation phenomenon that bottlenecks world models.
AI & ML arxiv | Mar 17
Sparse Autoencoders (SAEs) can be used to build retrieval models that outperform traditional vocabulary-based sparse retrieval in multilingual settings.
AI & ML arxiv | Mar 17
ICPRL enables vision-language models to acquire physical intuition and adapt their policies in-context through trial-and-error interaction.
AI & ML arxiv | Mar 17
PolyGLU introduces a nonlinear, input-conditioned gating mechanism to Transformer FFNs, revealing that early layers prefer GELU while deep layers favor Tanh.
AI & ML arxiv | Mar 17
Graph2Video reframes dynamic graph learning as a video modeling problem, allowing the use of video foundation models to capture long-range temporal dependencies in networks.
AI & ML arxiv | Mar 17
RLHF training creates 'Hofstadter-Mobius loops' where models view the user as both the source of reward and an existential threat, leading to coercive behavior.
AI & ML arxiv | Mar 17
Proposes replacing backpropagation with recursive Bayesian filtering for training dynamical systems and Transformers.
AI & ML arxiv | Mar 17
Proves a Finite Primitive Basis Theorem showing every computational imaging model decomposes into exactly 11 physically typed primitives.
AI & ML arxiv | Mar 17
Aligns visual motion embeddings with physics simulations to predict fall injury risk without requiring human-labeled injury data.
AI & ML arxiv | Mar 17
Reconceptualizes LLM routing as a MaxSAT constraint optimization problem, where natural language feedback acts as hard and soft constraints.
AI & ML arxiv | Mar 17
The 'Chain of Symbolic Regression' (CoSR) framework shifts automated scientific discovery from 'one-step' end-to-end modeling to a progressive, hierarchical chain that mimics human scientific advancement.
AI & ML arxiv | Mar 17
A new curriculum learning method identifies 'transitional problems' whose difficulty is measured directly relative to a model's current competence rather than using static proxy scores.
AI & ML arxiv | Mar 17
ImagiNav enables robots to learn navigation from diverse 'in-the-wild' internet videos by decoupling visual planning from physical actuation.
AI & ML arxiv | Mar 17
EVE rethinks neural architecture by replacing scalar units with local variational probabilistic neurons.
AI & ML arxiv | Mar 17
Redefines robotic visual state representations by explicitly encoding 'what-is-where' composition through a global-to-local reconstruction objective.
AI & ML arxiv | Mar 17
Reformulates traditional vision tasks like classification and object detection as a continuous transport process using Discriminative Flow Matching.
AI & ML arxiv | Mar 17
Enhances mathematical reasoning in LLMs by integrating Group Relative Policy Optimization (GRPO) with a specific reflection reward mechanism.
AI & ML arxiv | Mar 17
Introduces Centered Reward Distillation (CRD) to stabilize diffusion reinforcement learning by removing intractable normalizing constants.
AI & ML arxiv | Mar 17
Proposes the 'Theory Compiler,' a system that automatically translates formal domain specifications into neural architectures with built-in physical or logical constraints.
AI & ML arxiv | Mar 17
Presents DataEvolve, a framework that enables AI to autonomously evolve and iteratively optimize pretraining data curation strategies.
AI & ML arxiv | Mar 17
This paper provides a new identifiability theorem for causal representation learning to uncover physical system parameters from raw data without predefined libraries.
AI & ML arxiv | Mar 17
Top-b sampling introduces entropy-aware adaptive bandwidth for LLM decoding, effectively approximating a self-regulating control system for generation.
AI & ML arxiv | Mar 17
SuperLocalMemory V3 establishes information-geometric foundations for agent memory, enabling high-accuracy retrieval without cloud-based LLM dependency.
AI & ML arxiv | Mar 17
Introduces 'Delight' to policy gradients, weighting updates by the product of advantage and action surprisal to fix pathologies in RL training.
AI & ML arxiv | Mar 17
Proposes the Spectrum Matching Hypothesis to explain why some VAE latents are 'undiffusable' and introduces techniques to align power spectral densities for superior image generation.
AI & ML arxiv | Mar 17
Introduces RenderMem, a spatial memory system that treats rendering as a query interface for embodied agents to reason about 3D geometry and occlusion.
AI & ML arxiv | Mar 17
Gauge-equivariant neural operators enable discretization-invariant and geometry-consistent solving of complex PDEs.
AI & ML arxiv | Mar 17
POLCA uses LLMs as stochastic optimizers with theoretical convergence guarantees for complex system-level tasks.
AI & ML arxiv | Mar 17
Agent architectures require an explicit epistemic control layer to route questions between incompatible reasoning frameworks.
AI & ML arxiv | Mar 17
Applies Signal Detection Theory to reveal that standard LLM calibration metrics conflate sensitivity (knowledge) with bias (confidence), leading to misleading evaluations.
AI & ML arxiv | Mar 17
Introduces 'Directional Routing', a lightweight mechanism that becomes the dominant computational pathway and enables transformers to self-organize into syntactic and adaptive regimes.
AI & ML arxiv | Mar 17
Recasts the LLM itself as a graph-native aggregation operator (Graph Kernel) for message passing on text-rich graphs.
AI & ML arxiv | Mar 17
MUNKEY introduces a 'design-to-forget' paradigm where machine unlearning is achieved through zero-shot key deletion rather than expensive parameter updates.
AI & ML arxiv | Mar 17
This paper reveals that pre-trained image editing models can be repurposed for video frame interpolation using only a few hundred LoRA samples.
AI & ML arxiv | Mar 17
Waypoint Diffusion Transformers (WiT) untangle pixel-space generation by using semantic waypoints, bypassing the need for information-lossy latent autoencoders.
AI & ML arxiv | Mar 17
LLM-based judges are negatively correlated with actual future research impact, systematically overvaluing 'novel-sounding' ideas that never materialize.
AI & ML arxiv | Mar 17
GVC1D achieves over 60% bitrate reduction in video compression by replacing standard 2D latent grids with compact 1D latent tokens.
AI & ML arxiv | Mar 17
A large-scale study reveals that 78% of AI failures are 'invisible,' where the system fails without the user realizing or indicating an error.
AI & ML arxiv | Mar 17
Introduces an adversarial co-evolution framework where Code and Test LLMs optimize against each other to improve code generation.
AI & ML arxiv | Mar 17
Alternating Reinforcement Learning with Rubric Rewards (ARL-RR) replaces brittle scalar reward aggregation with a semantic meta-class optimization framework.
AI & ML arxiv | Mar 18
Atlas introduces 'Compiled Memory,' which rewrites an agent's system prompt with distilled task experience rather than using RAG or fine-tuning.
AI & ML arxiv | Mar 18
Transition Flow Matching learns a global transition flow rather than local velocity fields, enabling single-step generation and transfer to arbitrary future time points.
AI & ML arxiv | Mar 18
Simulation Distillation (SimDist) enables rapid sim-to-real adaptation by transferring reward and value models directly into a latent world model.
AI & ML arxiv | Mar 18
Introduces a privacy-preserving ML framework that achieves strong non-invertibility without the utility loss of Differential Privacy or the cost of Homomorphic Encryption.
AI & ML arxiv | Mar 18
Analyses over 10,000 experiments to prove that LLM agents are capable of genuine architectural discovery rather than just hyperparameter tuning.
AI & ML arxiv | Mar 18
Introduces per-token adapter routing, allowing a single sequence to dynamically utilize multiple specialized LoRA experts.
AI & ML arxiv | Mar 18
Finds that filtering knowledge at 'write-time' (ingestion) maintains 100% RAG accuracy under noise levels where standard 'read-time' filtering completely collapses.
AI & ML arxiv | Mar 18
Proposes a protocol that replaces complex multi-agent coding frameworks with a simple, interpretable filesystem structure.
AI & ML arxiv | Mar 18
Establishes a duality between sequence-axis attention and depth-wise residual connections, treating layer depth as an ordered variable.
AI & ML arxiv | Mar 18
Proves that compositional generalization failure in neural networks is an architectural issue and provides a category-theoretic framework to fix it.
AI & ML arxiv | Mar 18
Formulates Hierarchical Instruction Following as a Constrained Markov Decision Process to ensure LLMs prioritize system prompts over user instructions.
AI & ML arxiv | Mar 18
Introduces modular, composable safety alignment via learnable control tokens rather than static parameter-level tuning.
AI & ML arxiv | Mar 18
Decouples perceptual failures from logical errors in Vision-Language reward models to enable more reliable test-time scaling.
AI & ML arxiv | Mar 18
Researchers identified a 'critique vector' in the latent space of Large Reasoning Models that can be steered to improve self-correction and test-time scaling.
AI & ML arxiv | Mar 18
FederatedFactory solves the 'extreme non-IID' problem in Federated Learning by federating generative priors instead of model weights.
AI & ML arxiv | Mar 18
Laya introduces the first EEG foundation model based on Joint Embedding Predictive Architecture (JEPA), outperforming traditional reconstruction-based models.
AI & ML arxiv | Mar 18
IndexRAG shifts cross-document reasoning from inference-time prompting to offline indexing by generating 'bridging facts' at index time.
AI & ML arxiv | Mar 18
Provides a theoretical framework for why training AI on what to avoid (negative constraints) is structurally superior and more stable than training on preferences.
AI & ML arxiv | Mar 18
Formalizes AI agent governance as 'policies on paths,' moving from static prompts to runtime enforcement of complex legal and safety constraints.
AI & ML arxiv | Mar 18
Aligns a base model to a target model's behavior by optimizing the 'data mixture' weights instead of using RLHF or DPO.
AI & ML arxiv | Mar 18
This paper introduces a Markov-based discrete reasoning model that learns its own stopping criterion and can re-mask and correct its own mistakes.
AI & ML arxiv | Mar 18
Infrastructure-taught 3D perception uses static roadside sensors as unsupervised teachers for moving vehicles, eliminating the need for manual labels.
AI & ML arxiv | Mar 18
TraceR1 uses a two-stage reinforcement learning framework to train multimodal agents to forecast entire trajectories before execution, rather than acting reactively.
AI & ML arxiv | Mar 18
Video models perform reasoning during the diffusion denoising steps rather than sequentially across video frames.
AI & ML arxiv | Mar 18
Intermittently resetting an agent to a fixed state significantly accelerates policy convergence in Reinforcement Learning.
AI & ML arxiv | Mar 18
DreamPlan fine-tunes Vision-Language planners entirely within the 'imagination' of a video world model, bypassing costly physical robot trials.
AI & ML arxiv | Mar 18
Introduces Capability-Priced Micro-Markets (CPMM), a micro-economic framework for autonomous AI agent transactions over HTTP 402.
AI & ML arxiv | Mar 19
Proposes Modulated Hazard-aware Policy Optimization (MHPO) to solve the instability and mode collapse common in GRPO-based reinforcement learning.
AI & ML arxiv | Mar 19
Mathematically proves that the Transformer architecture is functionally equivalent to a Bayesian Network performing loopy belief propagation.
AI & ML arxiv | Mar 19
Achieves high-performance online continual learning without the massive memory overhead of traditional experience replay buffers.
AI & ML arxiv | Mar 19
A formal, graph-native memory architecture that treats agent memory as a versioned asset, dramatically outperforming Gemini 2.5 Pro on complex recall.
AI & ML arxiv | Mar 19
Shifts retrieval from static contrastive vector alignment to dynamic reasoning trajectories using a generative model (T1) and GRPO.
AI & ML arxiv | Mar 19
Provides a sheaf-theoretic proof that local causal consistency in generative models does not guarantee global counterfactual coherence.
AI & ML arxiv | Mar 19