Introduces secondary risks as a new class of LLM failures from benign prompts, defines two primitives, proposes SecLens search framework, and releases SecRiskBench showing risks are widespread across 16 models.
hub
Deep reinforcement learning: An overview
10 Pith papers cite this work. Polarity classification is still indexing.
10
Pith papers citing it
abstract
We give an overview of recent exciting achievements of deep reinforcement learning (RL). We discuss six core elements, six important mechanisms, and twelve applications. We start with background of machine learning, deep learning and reinforcement learning. Next we discuss core RL elements, including value function, in particular, Deep Q-Network (DQN), policy, reward, model, planning, and exploration. After that, we discuss important mechanisms for RL, including attention and memory, unsupervised learning, transfer learning, multi-agent RL, hierarchical RL, and learning to learn. Then we discuss various applications of RL, including games, in particular, AlphaGo, robotics, natural language processing, including dialogue systems, machine translation, and text generation, computer vision, neural architecture design, business management, finance, healthcare, Industry 4.0, smart grid, intelligent transportation systems, and computer systems. We mention topics not reviewed yet, and list a collection of RL resources. After presenting a brief summary, we close with discussions. Please see Deep Reinforcement Learning, arXiv:1810.06339, for a significant update.
hub tools
citation-role summary
background 3
citation-polarity summary
roles
background 3polarities
background 3representative citing papers
AACC combines online IOC for driving style identification with a Stackelberg game planner to proactively protect right-of-way against cut-ins, reporting up to 79.8% safety gains in simulation.
Seed-TTS models produce speech matching human naturalness and speaker similarity, with added controllability via self-distillation and reinforcement learning.
This is the first survey on vision-language-action models, providing a taxonomy across three lines, plus summaries of datasets, simulators, benchmarks, challenges, and future directions in embodied AI.
The paper introduces an inductive synthesis framework that generates verifiable deterministic program approximations of neural RL policies, preserving safety invariants via counterexample-guided search over state transition systems.
Agentic safety fails to generalize across tasks because the task-to-safe-controller mapping has a higher Lipschitz constant than the task-to-controller mapping alone, as proven in linear-quadratic control and demonstrated in quadcopter and LLM experiments.
ERL trains LLMs to erase faulty reasoning steps and regenerate them in place, yielding gains of up to 8.48% EM on multi-hop QA benchmarks like HotpotQA.
Cognitive models of user reasoning strategies with XAI methods on tabular data fit human forward-simulation decisions better than ML baselines and support hypothesis testing without new user studies.
The paper surveys the origins, frameworks, applications, and open challenges of AI agents built on large language models.
A deep RL architecture using imitation learning and reinforcement learning is proposed to model immediate and future values of search story recommendations in a Markov decision process framework.
citing papers explorer
-
Exploring the Secondary Risks of Large Language Models
Introduces secondary risks as a new class of LLM failures from benign prompts, defines two primitives, proposes SecLens search framework, and releases SecRiskBench showing risks are widespread across 16 models.
-
Anti-bullying Adaptive Cruise Control: A proactive right-of-way protection approach
AACC combines online IOC for driving style identification with a Stackelberg game planner to proactively protect right-of-way against cut-ins, reporting up to 79.8% safety gains in simulation.
-
Seed-TTS: A Family of High-Quality Versatile Speech Generation Models
Seed-TTS models produce speech matching human naturalness and speaker similarity, with added controllability via self-distillation and reinforcement learning.
-
A Survey on Vision-Language-Action Models for Embodied AI
This is the first survey on vision-language-action models, providing a taxonomy across three lines, plus summaries of datasets, simulators, benchmarks, challenges, and future directions in embodied AI.
-
An Inductive Synthesis Framework for Verifiable Reinforcement Learning
The paper introduces an inductive synthesis framework that generates verifiable deterministic program approximations of neural RL policies, preserving safety invariants via counterexample-guided search over state transition systems.
-
Why Does Agentic Safety Fail to Generalize Across Tasks?
Agentic safety fails to generalize across tasks because the task-to-safe-controller mapping has a higher Lipschitz constant than the task-to-controller mapping alone, as proven in linear-quadratic control and demonstrated in quadcopter and LLM experiments.
-
Erase to Improve: Erasable Reinforcement Learning for Search-Augmented LLMs
ERL trains LLMs to erase faulty reasoning steps and regenerate them in place, yielding gains of up to 8.48% EM on multi-hop QA benchmarks like HotpotQA.
-
CoAX: Cognitive-Oriented Attribution eXplanation User Model of Human Understanding of AI Explanations
Cognitive models of user reasoning strategies with XAI methods on tabular data fit human forward-simulation decisions better than ML baselines and support hypothesis testing without new user studies.
-
The Rise and Potential of Large Language Model Based Agents: A Survey
The paper surveys the origins, frameworks, applications, and open challenges of AI agents built on large language models.
-
Deep Reinforcement Learning for Personalized Search Story Recommendation
A deep RL architecture using imitation learning and reinforcement learning is proposed to model immediate and future values of search story recommendations in a Markov decision process framework.