SSL representations for echo LV segmentation and EF estimation recover substantially more task information with expressive downstream probes than with heuristic extraction, nearing supervised U-Net performance on EchoNet-Dynamic.
hub
arXiv preprint arXiv:2601.03220 , year=
19 Pith papers cite this work, alongside 1 external citations. Polarity classification is still indexing.
19
Pith papers citing it
1
external citations · external index
hub tools
years
2026 19representative citing papers
ICL-derived intrinsic rewards are biased in general MDPs but asymptotically match true learning progress in non-temporal settings, with supporting experiments.
Derives a resource-dependent real log canonical threshold for finite-time singular model selection via analytic effective potentials in non-equilibrium learning dynamics.
Creativity is defined as meta-learning where a frozen diffusion creator optimizes candidates for rapid improvement by an adapting appraiser such as an autoencoder or CLIP adapter.
A formal construction and synthetic experiment demonstrate that proxy rankings of pretraining datasets can reverse OOD accuracy rankings.
The paper introduces compositional interpretability as a category-theoretic framework that casts mechanistic explanations as commuting syntactic-semantic mappings optimized under faithfulness and complexity constraints derived from minimum description length.
Selecting 50% of robot demonstrations by maximizing exposure to reusable primitive-transition patterns outperforms full-data training while halving training steps.
The paper introduces an inductive generalization evaluation protocol for manipulation policies and shows that SOTA vision-language-action models fail on progressively harder task variants.
Distributional RL yields smoother objectives in chaotic systems because return distributions evolve more regularly than individual trajectories under the 1-Wasserstein metric.
Mixing cosine similarity with Rayleigh energy on a feature-space graph Laplacian improves head-tail coherence and modest semantic metrics over pure geometric retrieval.
TSD applies two physics metrics to identify salient trajectory segments for dataset compression and expansion in robotic imitation learning, yielding comparable performance with 25% less data on average.
Self-play RL regularized with 30 minutes of human data produces driving policies that coordinate with humans, training in 15 hours on one GPU with 2500x less data than imitation learning.
CAFE assesses the fit of observational CATE estimates by partitioning RCT data via propensity scores and comparing to experimental group averages, with theory and extensions for confounders.
The paper introduces an information-theoretic emergent communication framework for multi-agent task-solving in networking, deriving generalization bounds and validating on a hardware prototype.
SANEmerg enables emergent communication among bounded-intelligence AI agents for semantic-aware task fulfillment in AgentNet systems via a bandwidth-adaptable importance filter and MDL-based complexity regularizer.
Randomly masking a proposer's output vocabulary during training and generation sustains curriculum diversity and improves solver accuracy by +4.4 points at 8B in LLM co-evolution.
Open-ended intelligence is formalized as the compositional closure L(P,C) of primitives P under operators C, with next primitive prediction proposed as an objective to acquire reusable primitives and grammar for lifelong adaptation.
PPM injects parametric structural priors into generative models via a learnable mapping to improve probabilistic forecasts on non-stationary MTS data.
Interestingness is defined as an inductive signal for future compression progress, with proofs that expected progress decays exponentially with time since last breakthrough and that the Algorithmic Prior yields quadratic gains over the Length Prior.
citing papers explorer
-
Evaluating self-supervised echocardiographic representations across downstream extraction strategies for left-ventricular segmentation and ejection fraction estimation
SSL representations for echo LV segmentation and EF estimation recover substantially more task information with expressive downstream probes than with heuristic extraction, nearing supervised U-Net performance on EchoNet-Dynamic.
-
Can In-Context Learning Support Intrinsic Curiosity?
ICL-derived intrinsic rewards are biased in general MDPs but asymptotically match true learning progress in non-temporal settings, with supporting experiments.
-
Non-Equilibrium Model Selection via Finite-Time Thermodynamics
Derives a resource-dependent real log canonical threshold for finite-time singular model selection via analytic effective potentials in non-equilibrium learning dynamics.
-
Seeking the Unfamiliar but Memorable: Conceptual Creativity as Meta-Learning
Creativity is defined as meta-learning where a frozen diffusion creator optimizes candidates for rapid improvement by an adapting appraiser such as an autoencoder or CLIP adapter.
-
A Controlled Counterexample to Strong Proxy-Based Explanations of OOD Performance: in a Fixed Pretraining-and-Probing Setup
A formal construction and synthetic experiment demonstrate that proxy rankings of pretraining datasets can reverse OOD accuracy rankings.
-
From Mechanistic to Compositional Interpretability
The paper introduces compositional interpretability as a category-theoretic framework that casts mechanistic explanations as commuting syntactic-semantic mappings optimized under faithfulness and complexity constraints derived from minimum description length.
-
SIEVE: Structure-Aware Data Selection for Imitation Learning with VLA Models
Selecting 50% of robot demonstrations by maximizing exposure to reusable primitive-transition patterns outperforms full-data training while halving training steps.
-
Inductive Generalization for Robotic Manipulation
The paper introduces an inductive generalization evaluation protocol for manipulation policies and shows that SOTA vision-language-action models fail on progressively harder task variants.
-
On Distributional Reinforcement Learning in Chaotic Dynamical Systems
Distributional RL yields smoother objectives in chaotic systems because return distributions evolve more regularly than individual trajectories under the 1-Wasserstein metric.
-
From Embedding Geometry to Spectral Search: Energy Dispersion Networks For Vector Retrieval
Mixing cosine similarity with Rayleigh energy on a feature-space graph Laplacian improves head-tail coherence and modest semantic metrics over pure geometric retrieval.
-
TSD: A Physics-Inspired Trajectory Saliency Detector for Efficient Imitation Learning
TSD applies two physics metrics to identify salient trajectory segments for dataset compression and expansion in robotic imitation learning, yielding comparable performance with 25% less data on average.
-
Human-like autonomy emerges from self-play and a pinch of human data
Self-play RL regularized with 30 minutes of human data produces driving policies that coordinate with humans, training in 15 hours on one GPU with 2500x less data than imitation learning.
-
Assessing Estimate of CATE from Observational Data via an RCT Study
CAFE assesses the fit of observational CATE estimates by partitioning RCT data via propensity scores and comparing to experimental group averages, with theory and extensions for confounders.
-
Generalization Bounds of Emergent Communications for Agentic AI Networking
The paper introduces an information-theoretic emergent communication framework for multi-agent task-solving in networking, deriving generalization bounds and validating on a hardware prototype.
-
SANEmerg: An Emergent Communication Framework for Semantic-aware Agentic AI Networking
SANEmerg enables emergent communication among bounded-intelligence AI agents for semantic-aware task fulfillment in AgentNet systems via a bandwidth-adaptable importance filter and MDL-based complexity regularizer.
-
Vocabulary Dropout for Curriculum Diversity in LLM Co-Evolution
Randomly masking a proposer's output vocabulary during training and generation sustains curriculum diversity and improves solver accuracy by +4.4 points at 8B in LLM co-evolution.
-
A Compositional Framework for Open-ended Intelligence
Open-ended intelligence is formalized as the compositional closure L(P,C) of primitives P under operators C, with next primitive prediction proposed as an objective to acquire reusable primitives and grammar for lifelong adaptation.
-
Parametric Prior Mapping Framework for Non-stationary Probabilistic Time Series Forecasting
PPM injects parametric structural priors into generative models via a learnable mapping to improve probabilistic forecasts on non-stationary MTS data.
-
Interestingness as an Inductive Heuristic for Future Compression Progress
Interestingness is defined as an inductive signal for future compression progress, with proofs that expected progress decays exponentially with time since last breakthrough and that the Algorithmic Prior yields quadratic gains over the Length Prior.