RNG-Bench evaluates MLLMs on hidden-observation reconstruction in non-Markov games, finds forgetting as the dominant error source, and shows fine-tuning on optimal rollouts improves performance with transfer to other benchmarks.
hub
Textarena
14 Pith papers cite this work. Polarity classification is still indexing.
14
Pith papers citing it
hub tools
citation-role summary
background 2
dataset 1
citation-polarity summary
polarities
background 3representative citing papers
Steering language models with intermittent implicit trait reinforcements reduces misalignment contagion in multi-agent social dilemma games more effectively than system prompt repetition.
Terminal-Bench 2.0 is a new benchmark of 89 realistic terminal tasks on which frontier AI agents score below 65%.
Mindgames introduces a four-game evaluation platform for multi-agent LLM reasoning, runs a 944-agent competition, surfaces rule-adherence and error-survival limitations, and releases a 29k-game dataset with an offline scoring protocol.
DemoEvolve bootstraps harness evolution with demonstrations to achieve more stable and effective edits than self-rollout search in sparse-feedback environments like Balatro.
Presents Hack-Verifiable TextArena, a benchmark that embeds verifiable reward hacking opportunities into environments to enable deterministic measurement of exploitation by language models.
LLMs encode accurate but brittle internal beliefs about latent game states and convert them poorly into actions, creating systematic gaps that explain strategic failures.
This survey categorizes agentic environments for LLMs by eight attributes and domains, introduces symbolic and neural synthesis paradigms with evaluation, and outlines four agent evolution pathways plus three environment evolution paradigms.
The paper identifies four missing interfaces (data autolabelling, embodiment retargeting, physics-grounded world models, and video-based reward inference) as the central bottleneck beyond VLA scaling for robot intelligence.
The paper organizes research on generalist game AI into Dataset, Model, Harness, and Benchmark pillars and charts a five-level progression from single-game mastery to agents that create and live inside game multiverses.
DEPT detects training impasses in social language agents via dual-scale divergence and entropy, then uses asymmetric reshaping to restore exploration gradients and prevent policy homogenization.
Gym-V supplies 179 visual environments showing that observation scaffolding like captions and rules matters more for training success than the choice of RL algorithm.
A survey compiling RL methods, challenges, data resources, and applications for enhancing reasoning in large language models and large reasoning models since DeepSeek-R1.
citing papers explorer
-
Beyond the Current Observation: Evaluating Multimodal Large Language Models in Controllable Non-Markov Games
RNG-Bench evaluates MLLMs on hidden-observation reconstruction in non-Markov games, finds forgetting as the dominant error source, and shows fine-tuning on optimal rollouts improves performance with transfer to other benchmarks.
-
Mitigating Misalignment Contagion by Steering with Implicit Traits
Steering language models with intermittent implicit trait reinforcements reduces misalignment contagion in multi-agent social dilemma games more effectively than system prompt repetition.
-
Terminal-Bench: Benchmarking Agents on Hard, Realistic Tasks in Command Line Interfaces
Terminal-Bench 2.0 is a new benchmark of 89 realistic terminal tasks on which frontier AI agents score below 65%.
-
MINDGAMES: A Live Arena for Evaluating Social and Strategic Reasoning in Multi-Agent LLMs
Mindgames introduces a four-game evaluation platform for multi-agent LLM reasoning, runs a 944-agent competition, surfaces rule-adherence and error-survival limitations, and releases a 29k-game dataset with an offline scoring protocol.
-
DemoEvolve: Overcoming Sparse Feedback in Agentic Harness Evolution with Demonstrations
DemoEvolve bootstraps harness evolution with demonstrations to achieve more stable and effective edits than self-rollout search in sparse-feedback environments like Balatro.
-
Hack-Verifiable Environments: Towards Evaluating Reward Hacking at Scale
Presents Hack-Verifiable TextArena, a benchmark that embeds verifiable reward hacking opportunities into environments to enable deterministic measurement of exploitation by language models.
-
Why Do LLMs Struggle in Strategic Play? Broken Links Between Observations, Beliefs, and Actions
LLMs encode accurate but brittle internal beliefs about latent game states and convert them poorly into actions, creating systematic gaps that explain strategic failures.
-
Agentic Environment Engineering for Large Language Models: A Survey of Environment Modeling, Synthesis, Evaluation, and Application
This survey categorizes agentic environments for LLMs by eight attributes and domains, introduces symbolic and neural synthesis paradigms with evaluation, and outlines four agent evolution pathways plus three environment evolution paradigms.
-
Robots Need More than VLA and World Models
The paper identifies four missing interfaces (data autolabelling, embodiment retargeting, physics-grounded world models, and video-based reward inference) as the central bottleneck beyond VLA scaling for robot intelligence.
-
Towards Generalist Game Players: An Investigation of Foundation Models in the Game Multiverse
The paper organizes research on generalist game AI into Dataset, Model, Harness, and Benchmark pillars and charts a five-level progression from single-game mastery to agents that create and live inside game multiverses.
-
Breaking the Impasse: Dual-Scale Evolutionary Policy Training for Social Language Agents
DEPT detects training impasses in social language agents via dual-scale divergence and entropy, then uses asymmetric reshaping to restore exploration gradients and prevent policy homogenization.
-
Gym-V: A Unified Vision Environment System for Agentic Vision Research
Gym-V supplies 179 visual environments showing that observation scaffolding like captions and rules matters more for training success than the choice of RL algorithm.
-
A Survey of Reinforcement Learning for Large Reasoning Models
A survey compiling RL methods, challenges, data resources, and applications for enhancing reasoning in large language models and large reasoning models since DeepSeek-R1.
- When Context Returns: Toward Robust Internalization in On-Policy Distillation