MUSE shows that the native timestep embedding in diffusion models acts as a parameter-free steering signal for multi-task monocular depth and normal estimation via manifold decoupling in latent space.
hub
arXiv preprint arXiv:2009.09796 (2020)
15 Pith papers cite this work. Polarity classification is still indexing.
15
Pith papers citing it
hub tools
citation-role summary
background 1
citation-polarity summary
verdicts
UNVERDICTED 15roles
background 1polarities
background 1representative citing papers
4DLidarOpen is a new open dataset providing synchronized 4D FMCW Lidar velocity measurements, multi-Lidar and camera data, and 3D bounding-box annotations with track IDs to support benchmarks on 3D detection, BEV segmentation, flow prediction, and motion forecasting.
CODI compresses explicit CoT into continuous space via self-distillation and is the first implicit method to match explicit CoT performance on GSM8k at GPT-2 scale with 3.1x compression and 28.2% higher accuracy than prior implicit approaches.
FAROS uses flow-guided propagation from zero-shot masks and optical flow to create dense temporally consistent labels from sparse keyframes, improving joint multi-task learning across temporal and spatial surgical tasks on GraSP, MISAW, and AutoLaparo.
Everywhere learning trains AI to meet pointwise loss constraints almost surely, backed by approximate duality theory for generalization and L1 regularization on relaxations.
SPLIT-PINN infers drift fields in Liouville transport equations from data using marginal corrections and orthogonality constraints to enable probabilistic predictions of microstructural evolution across polycrystal realizations.
Octopus introduces history-free gradient orthogonalization in a two-stage finetuning framework to achieve state-of-the-art continual learning results for multimodal LLMs on the UCIT benchmark.
USTri is a tri-stage ultrasound system that trains a generalist model, fine-tunes specialists while frozen, and deploys an agent for workflow orchestration, claiming top performance across 4 task types and 27 datasets.
FedRouter clusters adapters locally per task samples and globally across clients to create task-centric personalized models, improving generalization and reducing task interference in federated fine-tuning.
Introduces progressive task-specific multi-task adaptation for vision transformers, sharing adapters early and specializing later with gradient-based task allocation, outperforming prior methods on PASCAL and NYUD-v2 with fewer trainable parameters.
A multi-head private-layer ensemble in a unified multi-task setup reaches 89.96% accuracy on 453K e-commerce examples and improves low-resource tasks by up to 14% while revealing encoder-decoder asymmetry in task identity handling.
Mechanistic analysis of LLMs finds partial overlap in math-associated parameters across languages, concentrated in middle layers, with systematic language-dependent differences.
Contrastive KERMT pretraining on molecular graphs yields 7.6-9.9% average gains over KERMT baseline on Biogen, ExpansionRX, and ChEMBL-MT ADME endpoints via a single probabilistic latent-variable objective and task-specific GNN heads.
Derives tighter generalization bounds for vector-valued neural networks and deep kernel methods in multi-task learning via Koopman and PF operators, with sketching for efficiency and a new vvRKHS framework.
MacroNav learns multi-scale navigation-centric representations through multi-task self-supervised learning and combines them with graph-based reinforcement learning for efficient action selection, reporting gains in success rate and path efficiency over prior methods.
citing papers explorer
-
MUSE: Unlocking Timestep as Native Task Steering for One-Step Dense Prediction
MUSE shows that the native timestep embedding in diffusion models acts as a parameter-free steering signal for multi-task monocular depth and normal estimation via manifold decoupling in latent space.
-
4DLidarOpen: An Open 4D FMCW Lidar Dataset for Motion-Aware Autonomous Driving
4DLidarOpen is a new open dataset providing synchronized 4D FMCW Lidar velocity measurements, multi-Lidar and camera data, and 3D bounding-box annotations with track IDs to support benchmarks on 3D detection, BEV segmentation, flow prediction, and motion forecasting.
-
CODI: Compressing Chain-of-Thought into Continuous Space via Self-Distillation
CODI compresses explicit CoT into continuous space via self-distillation and is the first implicit method to match explicit CoT performance on GSM8k at GPT-2 scale with 3.1x compression and 28.2% higher accuracy than prior implicit approaches.
-
Temporally Consistent Label Interpolation for Robust Surgical Multi-Task Learning under Challenging Conditions
FAROS uses flow-guided propagation from zero-shot masks and optical flow to create dense temporally consistent labels from sparse keyframes, improving joint multi-task learning across temporal and spatial surgical tasks on GraSP, MISAW, and AutoLaparo.
-
Everywhere Learning: Artificial Intelligence with Pointwise Constraints
Everywhere learning trains AI to meet pointwise loss constraints almost surely, backed by approximate duality theory for generalization and L1 regularization on relaxations.
-
SPLIT-PINN: Separable Probability Learning Technique via Physics-Informed Neural Networks for High-Dimensional Probabilistic Modeling
SPLIT-PINN infers drift fields in Liouville transport equations from data using marginal corrections and orthogonality constraints to enable probabilistic predictions of microstructural evolution across polycrystal realizations.
-
Octopus: History-Free Gradient Orthogonalization for Continual Learning in Multimodal Large Language Models
Octopus introduces history-free gradient orthogonalization in a two-stage finetuning framework to achieve state-of-the-art continual learning results for multimodal LLMs on the UCIT benchmark.
-
Unified Ultrasound Intelligence Toward an End-to-End Agentic System
USTri is a tri-stage ultrasound system that trains a generalist model, fine-tunes specialists while frozen, and deploys an agent for workflow orchestration, claiming top performance across 4 task types and 27 datasets.
-
Task-Centric Personalized Federated Fine-Tuning of Language Models
FedRouter clusters adapters locally per task samples and globally across clients to create task-centric personalized models, improving generalization and reducing task interference in federated fine-tuning.
-
Parameter-Efficient Multi-Task Learning via Progressive Task-Specific Adaptation
Introduces progressive task-specific multi-task adaptation for vision transformers, sharing adapters early and specializing later with gradient-based task allocation, outperforming prior methods on PASCAL and NYUD-v2 with fewer trainable parameters.
-
Unified Multi-Task Relevance Modeling for E-Commerce: Comparing Task Routing Architectures Across LLMs and Cross-Encoders
A multi-head private-layer ensemble in a unified multi-task setup reaches 89.96% accuracy on 453K e-commerce examples and improves low-resource tasks by up to 14% while revealing encoder-decoder asymmetry in task identity handling.
-
LLM Parameters for Math Across Languages: Shared or Separate?
Mechanistic analysis of LLMs finds partial overlap in math-associated parameters across languages, concentrated in middle layers, with systematic language-dependent differences.
-
Probabilistic Contrastive Pretraining for Multi-task ADME Property Prediction
Contrastive KERMT pretraining on molecular graphs yields 7.6-9.9% average gains over KERMT baseline on Biogen, ExpansionRX, and ChEMBL-MT ADME endpoints via a single probabilistic latent-variable objective and task-specific GNN heads.
-
Operator-Based Generalization Bound for Deep Learning: Insights on Multi-Task Learning
Derives tighter generalization bounds for vector-valued neural networks and deep kernel methods in multi-task learning via Koopman and PF operators, with sketching for efficiency and a new vvRKHS framework.
-
MacroNav: Multi-Task Context Representation Learning Enables Efficient Navigation in Unknown Environments
MacroNav learns multi-scale navigation-centric representations through multi-task self-supervised learning and combines them with graph-based reinforcement learning for efficient action selection, reporting gains in success rate and path efficiency over prior methods.