The paper gives the first tight necessity and sufficiency conditions for successful reward poisoning attacks in linear MDPs.
Canonical reference
Title resolution pending
Canonical reference. 100% of citing Pith papers cite this work as background.
17
Pith papers citing it
Background
100% of classified citations
citation-role summary
background 5
citation-polarity summary
roles
background 5polarities
background 5representative citing papers
Adam-HNAG is a splitting-based reformulation of Adam that yields the first convergence proof for Adam-type methods, including accelerated rates, in convex smooth optimization.
LLM-generated heuristics for HTN planning nearly match PANDA planner coverage while reducing search effort on 83% of shared problems across six benchmark domains.
FedQual improves federated label distribution learning under heterogeneous annotation quality via quality-adaptive training with a global anchor and reliability-aware aggregation, backed by new benchmarks and a proof that client-specific calibration strictly outperforms uniform calibration.
ABox abduction under repair semantics for inconsistent KBs yields a full complexity landscape in lightweight description logics DL-Lite and EL_bot.
DLM4G applies graph-aware adaptive noising in a diffusion framework to generate text from graphs, outperforming larger autoregressive and diffusion baselines in factual grounding and edit sensitivity on three datasets plus molecule captioning.
LoRA gradient descent converges to a stationary point at rate O(1/log T).
MLHCA is a new ML-powered combinatorial auction combining value and demand queries to reduce efficiency loss by up to 10x and queries by up to 58% versus prior SOTA.
UHD-GCN-BIQA models structural dependencies among sampled patches via a hybrid kNN graph and residual graph convolutions to achieve competitive PLCC and SRCC with the lowest RMSE on the UHD-IQA benchmark for blind ultra-high-definition image quality assessment.
Derives expectation consistency condition as necessary and sufficient for calibration under covariate shift and proposes ECL loss with matching sample complexity to ECE.
VISOR is a VLM-based automated test oracle that evaluates robot task correctness and quality from videos while reporting its own uncertainty, tested on GPT and Gemini across four tasks and over 1000 videos with Gemini showing higher recall and GPT higher precision but low uncertainty-correctness tie
Random slicing for subsampling combined with Nadaraya-Watson smoothing enables faster and improved persistence-based topological optimization of point clouds in 2D and 3D.
SACHI enriches agent representations via graph transformer convolutions over inter-agent graphs to enable holistic information integration, outperforming baselines across five cooperative tasks with statistical significance.
SHINE trains a scalable in-context hypernetwork to generate high-quality LoRA adapters from contexts in one pass, enabling efficient LLM adaptation that saves time and compute compared to standard fine-tuning.
Cognitive forcing interventions reduce overreliance on AI recommendations more than simple explanations, with effects moderated by individual need for cognition.
ResGIN-Att predicts drug synergy by extracting multi-scale molecular features with residual GIN, fusing them via LSTM, and modeling interactions with cross-attention, achieving competitive results on five benchmark datasets.
A survey of trajectory prediction techniques for autonomous vehicles that proposes a taxonomy, overviews the prediction pipeline, and highlights remaining research gaps.
citing papers explorer
-
When Can You Poison Rewards? A Tight Characterization of Reward Poisoning in Linear MDPs
The paper gives the first tight necessity and sufficiency conditions for successful reward poisoning attacks in linear MDPs.
-
Adam-HNAG: A Convergent Reformulation of Adam with Accelerated Rate
Adam-HNAG is a splitting-based reformulation of Adam that yields the first convergence proof for Adam-type methods, including accelerated rates, in convex smooth optimization.
-
Hierarchical Task Network Planning with LLM-Generated Heuristics
LLM-generated heuristics for HTN planning nearly match PANDA planner coverage while reducing search effort on 83% of shared problems across six benchmark domains.
-
Trustworthy Federated Label Distribution Learning under Annotation Quality Disparity
FedQual improves federated label distribution learning under heterogeneous annotation quality via quality-adaptive training with a global anchor and reliability-aware aggregation, backed by new benchmarks and a proof that client-specific calibration strictly outperforms uniform calibration.
-
ABox Abduction for Inconsistent Knowledge Bases under Repair Semantics
ABox abduction under repair semantics for inconsistent KBs yields a full complexity landscape in lightweight description logics DL-Lite and EL_bot.
-
Factual and Edit-Sensitive Graph-to-Sequence Generation via Graph-Aware Adaptive Noising
DLM4G applies graph-aware adaptive noising in a diffusion framework to generate text from graphs, outperforming larger autoregressive and diffusion baselines in factual grounding and edit sensitivity on three datasets plus molecule captioning.
-
On the Convergence Rate of LoRA Gradient Descent
LoRA gradient descent converges to a stationary point at rate O(1/log T).
-
Prices, Bids, Values: One ML-Powered Combinatorial Auction to Rule Them All
MLHCA is a new ML-powered combinatorial auction combining value and demand queries to reduce efficiency loss by up to 10x and queries by up to 58% versus prior SOTA.
-
Ultra-High-Definition Image Quality Assessment via Graph Representation Learning
UHD-GCN-BIQA models structural dependencies among sampled patches via a hybrid kNN graph and residual graph convolutions to achieve competitive PLCC and SRCC with the lowest RMSE on the UHD-IQA benchmark for blind ultra-high-definition image quality assessment.
-
Expectation Consistency Loss: Rethink Confidence Calibration under Covariate Shift
Derives expectation consistency condition as necessary and sufficient for calibration under covariate shift and proposes ECL loss with matching sample complexity to ECE.
-
VISOR: A Vision-Language Model-based Test Oracle for Testing Robots
VISOR is a VLM-based automated test oracle that evaluates robot task correctness and quality from videos while reporting its own uncertainty, tested on GPT and Gemini across four tasks and over 1000 videos with Gemini showing higher recall and GPT higher precision but low uncertainty-correctness tie
-
Towards Scalable Persistence-Based Topological Optimization
Random slicing for subsampling combined with Nadaraya-Watson smoothing enables faster and improved persistence-based topological optimization of point clouds in 2D and 3D.
-
SACHI: Structured Agent Coordination via Holistic Information Integration in Multi-Agent Reinforcement Learning
SACHI enriches agent representations via graph transformer convolutions over inter-agent graphs to enable holistic information integration, outperforming baselines across five cooperative tasks with statistical significance.
-
SHINE: A Scalable In-Context Hypernetwork for Mapping Context to LoRA in a Single Pass
SHINE trains a scalable in-context hypernetwork to generate high-quality LoRA adapters from contexts in one pass, enabling efficient LLM adaptation that saves time and compute compared to standard fine-tuning.
-
To Trust or to Think: Cognitive Forcing Functions Can Reduce Overreliance on AI in AI-assisted Decision-making
Cognitive forcing interventions reduce overreliance on AI recommendations more than simple explanations, with effects moderated by individual need for cognition.
-
Drug Synergy Prediction via Residual Graph Isomorphism Networks and Attention Mechanisms
ResGIN-Att predicts drug synergy by extracting multi-scale molecular features with residual GIN, fusing them via LSTM, and modeling interactions with cross-attention, achieving competitive results on five benchmark datasets.
-
Trajectory Prediction for Autonomous Driving: Progress, Limitations, and Future Directions
A survey of trajectory prediction techniques for autonomous vehicles that proposes a taxonomy, overviews the prediction pipeline, and highlights remaining research gaps.