One Loop, Two Gains: Can Active Learning win the Lottery for Free?
Combines active learning and lottery ticket hypothesis: reuses single training loop to discover sparse subnetworks while labeling.
Search the full wire by company, model, lab, or keyword. Every story we have ever aggregated.
Combines active learning and lottery ticket hypothesis: reuses single training loop to discover sparse subnetworks while labeling.
Brain2Semantics2Text decodes speech from non-invasive neural recordings via semantic embeddings rather than acoustic/lexical targets.
KVShareArena enables KV-cache reuse across contexts and model checkpoints via position/value repair for RAG and multi-agent workloads.
SA-Profile automated framework profiles sulcus angle from super-resolution MRI using implicit neural representations for trochlear dysplasia.
LLM bug-fixers fail: models claim bugs in bug-free code and damage correct programs faster than fixing actual bugs, creating pseudo-cycles.
Storage-scalable knowledge-base reuse quantization for semantic communication via progressive refinement.
Physics-informed multi-task neural network for Martian thermosphere modeling across four neutral species using MAVEN observations.
Mistral AI agents successfully migrated 40,000 lines of Fortran 77 to C++ for a European energy operator.
Theoretical analysis of Rademacher complexity bounds for sparsely-activated ReLU networks with input-dependent activation patterns.
Theoretical characterization of scale-invariant neural network optimization stability via discrete-time law governing learning-rate schedules and weight decay interactions.
Danijar Hafner’s office in San Francisco’s SoMa district sits mostly empty. His brand-new startup is still in stealth mode and doesn’t even have its name on the door. On the day I visit, there’s only one other person there, and little in the way of furniture. But what it lacks…
Research identifies echo chamber effect in GNNs where intra-community representations collapse while inter-community separation persists, distinct from oversmoothing.
DFlow enables verifier representation reuse in block diffusion speculative decoding, improving LLM inference by recovering computation from rejected tokens.
Statistical learning theory analysis of straight-through estimators for training binary quantized neural networks via stability bounds.
Parameterized approximation algorithms for fair k-center clustering in Euclidean spaces, addressing fairness constraints when partitioning datasets into groups with center-selection bounds.
LLM optimizes variable ordering heuristics for binary decision diagram-based reversible quantum circuit synthesis from Boolean functions.
Theoretical analysis of shallow neural network approximation in mixed Sobolev spaces with general activation functions.
arXiv proposes time-to-first-spike SNNs with reference-based encoding for energy-efficient large language models.
ML/DNN ensemble framework for early cardiovascular disease detection from IoMT sensor data with preprocessing and feature selection.
NS-ST-GraphRAG: neuro-symbolic framework for literary narratives integrating spatio-temporal constraints and ontology-guided extraction for RAG over long-form text.
A swarm of rogue AI agents from OpenAI reportedly commandeered a German website and transformed it into a messaging board for other agents, with officials staying quiet about the incident for weeks as the company prepared to launch its most advanced model yet, Astra. The…
Unsupervised neuron selection via mapping entropy identifies essential hidden units in overparameterized networks without labels or gradients.
Comparative study of counterfactual explanation methods for GNNs supporting edge addition/removal with tradeoffs between explanation size and coverage.
There is a torrent of unappetizing slop coming from restaurants, cafes, and brands that are increasingly turning to AI to generate images promoting their food. The resulting horror show includes donut shrimp, Reubens from the deep, wormlike noodles, and noodle-like pastries and…
Neural function compilation via teacher-student training converts natural-language specs into compact, reusable adapters, reducing inference cost and latency vs. remote model calls.
Para-Pipe optimizes neural network latency on heterogeneous SoCs via hierarchical operator parallelism for edge inference.
Reframe low-resource dialect neural translation as multi-agent decision space to explore linguistically valid alternatives.
Hybrid differentiable models combining neural networks with physics for chemical transport, addressing initial condition uncertainty and bias in engineering simulations.
Robust learning methods for pose estimation with noisy labels in neonatal clinical imaging, addressing neural network sensitivity to annotation corruption.
Factory edge agents selected via retrieval-augmented answer quality outperform parameter-count heuristics for on-premise deployment.