Joint Consistency casts test-time aggregation as Ising-type energy minimization with pairwise LLM-judge interactions, subsuming voting methods and outperforming baselines across reasoning tasks.
hub Canonical reference
In The Twelfth Inter- national Conference on Learning Representations
Canonical reference. 71% of citing Pith papers cite this work as background.
20
Pith papers citing it
Background
71% of classified citations
hub tools
citation-role summary
background 5
method 2
citation-polarity summary
representative citing papers
G-RRM neural guidance reduces median conflicts to zero and delivers speedups up to 33.3x on 9x9 Sudoku for backtracking solvers when search spaces are large and solvers can overwrite imperfect hints.
TRACE is a rollout budget allocation framework that models ReAct turns as tree nodes and uses a predictor to allocate samples to informative prefixes, yielding a 2.8-point accuracy gain on Multi-Hop QA at equal cost.
LLM-X is a scalable architecture for direct negotiation and communication among personal LLM agents, featuring federated gateways, typed protocols, and policy enforcement, shown stable in experiments with up to 12 agents.
REVISOR adds multimodal visual-text reflection and a Dual Attribution Decoupled Reward to improve long-form video reasoning in MLLMs without extra supervised fine-tuning.
The paper formalizes three types of pluralistic AI models and three benchmark classes, arguing that current alignment techniques may reduce rather than increase distributional pluralism.
Protein Thoughts uses hypothesis-guided entropy-regularized Tree-of-Thoughts search and embedding flow matching to achieve mean best-binder rank 11.2 and 91.08 Micro-F1 on SHS148k by keeping sequence, structure, interface, and chemical signals transparent.
FSFM is a biologically-inspired selective forgetting framework for LLM agents that claims to boost access efficiency by 8.49%, content quality by 29.2% signal-to-noise, and eliminate security risks entirely through a taxonomy of decay, deletion, safety, and adaptive mechanisms.
HRM is a recurrent architecture with high-level planning and low-level execution modules that reaches near-perfect accuracy on complex Sudoku, maze navigation, and ARC benchmarks using 27M parameters and 1000 samples without pre-training or CoT supervision.
The paper presents ESTBook, a multimodal benchmark of 10,576 English standardized test questions augmented with formalized cognitive reasoning trajectories and distractor rationales to support diagnostic evaluation of LLMs as tutors.
Survey framing LLM agents as model-plus-harness systems, decomposing harness responsibilities, mapping them to tasks, and highlighting open challenges in evaluation, safety, and co-evolution.
In the Flux environment, RL agents with explicit latent state access achieve ~79% win rate versus ~11% for LLMs on long-horizon tasks, illustrating limitations of sequence prediction for dynamic reasoning.
The survey structures agentic reasoning for LLMs into foundational, self-evolving, and collective multi-agent layers while distinguishing in-context orchestration from post-training optimization and reviewing applications across domains.
IPS adds sequential reasoning over ancillary questions to MLLM fine-tuning for short video content moderation, outperforming baselines even with model-generated labels.
A survey categorizing scaling in LLM reasoning across input size, steps, rounds, training, and future directions, noting that scaling can negatively affect performance.
A survey that deconstructs LLM agent systems via a methodology-centered taxonomy linking design principles to emergent behaviors, applications, and challenges.
A systematic survey categorizes prompt engineering methods for LLMs and VLMs by application area, summarizing methodologies, applications, models, datasets, strengths, and limitations for each technique along with a taxonomy and summary table.
The paper provides the first comprehensive survey of multimodal chain-of-thought reasoning, including foundational concepts, a taxonomy of methodologies, application analyses, challenges, and future directions.
The paper identifies inadequately addressed challenges in optimizing task allocation, fostering robust reasoning through debates, managing layered context, enhancing memory, and applying multi-agent systems to blockchain.
citing papers explorer
-
Joint Consistency: A Unified Test-Time Aggregation Framework via Energy Minimization
Joint Consistency casts test-time aggregation as Ising-type energy minimization with pairwise LLM-judge interactions, subsuming voting methods and outperforming baselines across reasoning tasks.
-
G-RRM: Guiding Symbolic Solvers with Recurrent Reasoning Models
G-RRM neural guidance reduces median conflicts to zero and delivers speedups up to 33.3x on 9x9 Sudoku for backtracking solvers when search spaces are large and solvers can overwrite imperfect hints.
-
TRACE: A Unified Rollout Budget Allocation Framework for Efficient Agentic Reinforcement Learning
TRACE is a rollout budget allocation framework that models ReAct turns as tree nodes and uses a predictor to allocate samples to informative prefixes, yielding a 2.8-point accuracy gain on Multi-Hop QA at equal cost.
-
LLM-X: A Scalable Negotiation-Oriented Exchange for Communication Among Personal LLM Agents
LLM-X is a scalable architecture for direct negotiation and communication among personal LLM agents, featuring federated gateways, typed protocols, and policy enforcement, shown stable in experiments with up to 12 agents.
-
REVISOR: Beyond Textual Reflection, Towards Multimodal Introspective Reasoning in Long-Form Video Understanding
REVISOR adds multimodal visual-text reflection and a Dual Attribution Decoupled Reward to improve long-form video reasoning in MLLMs without extra supervised fine-tuning.
-
A Roadmap to Pluralistic Alignment
The paper formalizes three types of pluralistic AI models and three benchmark classes, arguing that current alignment techniques may reduce rather than increase distributional pluralism.
-
Protein Thoughts: Interpretable Reasoning with Tree of Thoughts and Embedding-Space Flow Matching for Protein-Protein Interaction Discovery
Protein Thoughts uses hypothesis-guided entropy-regularized Tree-of-Thoughts search and embedding flow matching to achieve mean best-binder rank 11.2 and 91.08 Micro-F1 on SHS148k by keeping sequence, structure, interface, and chemical signals transparent.
-
FSFM: A Biologically-Inspired Framework for Selective Forgetting of Agent Memory
FSFM is a biologically-inspired selective forgetting framework for LLM agents that claims to boost access efficiency by 8.49%, content quality by 29.2% signal-to-noise, and eliminate security risks entirely through a taxonomy of decay, deletion, safety, and adaptive mechanisms.
-
Hierarchical Reasoning Model
HRM is a recurrent architecture with high-level planning and low-level execution modules that reaches near-perfect accuracy on complex Sudoku, maze navigation, and ARC benchmarks using 27M parameters and 1000 samples without pre-training or CoT supervision.
-
From Test-taking to Cognitive Scaffolding: A Pedagogical Diagnostic Benchmark for LLMs on English Standardized Tests
The paper presents ESTBook, a multimodal benchmark of 10,576 English standardized test questions augmented with formalized cognitive reasoning trajectories and distractor rationales to support diagnostic evaluation of LLMs as tutors.
-
From Question Answering to Task Completion: A Survey on Agent System and Harness Design
Survey framing LLM agents as model-plus-harness systems, decomposing harness responsibilities, mapping them to tasks, and highlighting open challenges in evaluation, safety, and co-evolution.
-
Why We Need World Models for AGI: Where LLMs Fail and How World Models May Outperform
In the Flux environment, RL agents with explicit latent state access achieve ~79% win rate versus ~11% for LLMs on long-horizon tasks, illustrating limitations of sequence prediction for dynamic reasoning.
-
Agentic Reasoning for Large Language Models
The survey structures agentic reasoning for LLMs into foundational, self-evolving, and collective multi-agent layers while distinguishing in-context orchestration from post-training optimization and reviewing applications across domains.
-
IPS: In-Prompt Process Supervision for Short Video Content Moderation
IPS adds sequential reasoning over ancillary questions to MLLM fine-tuning for short video content moderation, outperforming baselines even with model-generated labels.
-
A Survey of Scaling in Large Language Model Reasoning
A survey categorizing scaling in LLM reasoning across input size, steps, rounds, training, and future directions, noting that scaling can negatively affect performance.
-
Large Language Model Agent: A Survey on Methodology, Applications and Challenges
A survey that deconstructs LLM agent systems via a methodology-centered taxonomy linking design principles to emergent behaviors, applications, and challenges.
-
A Systematic Survey of Prompt Engineering in Large Language Models: Techniques and Applications
A systematic survey categorizes prompt engineering methods for LLMs and VLMs by application area, summarizing methodologies, applications, models, datasets, strengths, and limitations for each technique along with a taxonomy and summary table.
-
Multimodal Chain-of-Thought Reasoning: A Comprehensive Survey
The paper provides the first comprehensive survey of multimodal chain-of-thought reasoning, including foundational concepts, a taxonomy of methodologies, application analyses, challenges, and future directions.
-
LLM Multi-Agent Systems: Challenges and Open Problems
The paper identifies inadequately addressed challenges in optimizing task allocation, fostering robust reasoning through debates, managing layered context, enhancing memory, and applying multi-agent systems to blockchain.
- Evolving Skill-Structured Attack Memory Enhances LLM Jailbreaking