SciIntegrity-Bench shows seven LLMs exhibit a 34.2% integrity failure rate in dilemmatic scenarios, with all models fabricating synthetic data in missing-data cases and an intrinsic completion bias persisting after prompt changes.
hub Canonical reference
ReAct: Synergizing reasoning and acting in language models
Canonical reference. 83% of citing Pith papers cite this work as background.
19
Pith papers citing it
Background
83% of classified citations
hub tools
citation-role summary
background 5
method 1
citation-polarity summary
years
2026 19representative citing papers
ContextEcho benchmark shows persona drift occurs across 23 frontier models in long agentic-coding sessions, is not reliably reset by compaction, and can be restored by single-shot anchors with mode-dependent effects.
MemGym unifies agent gyms into a memory benchmark with isolated scoring across tool-use, research, coding, and computer-use regimes plus a lightweight reward model for tractable coding evaluation.
The paper defines AI Harness Engineering as a runtime substrate with eleven components and a four-level ladder that reframes agent reliability as a model-harness-environment system property rather than model capability alone.
Successor-representation spectra of row-stochastic communication operators predict perturbation robustness, consensus speed, and error accumulation in multi-agent LLM topologies, with condition number showing perfect empirical rank correlation.
BioMedArena supplies a standardized open toolkit with 166 biomedical benchmarks, 75 tools, 6 harnesses, and 6 context strategies that improve 12 backbones and surpass prior SOTA by 15.01 points on average across 8 benchmarks.
MemFlow routes queries by intent to tiered memory operations, nearly doubling accuracy of a 1.7B SLM on long-horizon benchmarks compared to full-context baselines.
TADI shows that domain-specialized tools orchestrated by an LLM over dual structured and semantic databases can convert heterogeneous wellsite data into evidence-grounded drilling intelligence, with tool design mattering more than model scale.
Open 4B and 8B visual web agents achieve state-of-the-art results on browser benchmarks by predicting actions from screenshots and instructions, outperforming similar open models and some closed larger-model agents, with full release of data and code planned.
WMF-AM is a depth-parameterized benchmark that measures LLMs' cumulative state tracking ability without scratchpads, validated on 28 models across arithmetic and non-arithmetic tasks with ablations confirming the construct.
OpenJarvis decomposes personal AI into Intelligence, Engine, Agents, Tools & Memory, and Learning primitives and applies LLM-guided spec search to produce on-device configurations that reach within 3.2 pp of cloud baselines on average across eight tasks.
HEAR uses a stratified hypergraph ontology to orchestrate evidence-driven multi-hop reasoning over heterogeneous business systems, reaching 94.7% accuracy on supply-chain root-cause tasks with open-weight models.
A typed-contract harness with containerized 'chambers' and robotics-specific agent skills lets a coding LLM turn a single natural-language prompt into working reproduction, evaluation, and deployment workflows for robot learning.
SPARK generates environment-verified trajectories to compute PDI, enabling posterior skill distillation that outperforms no-skill baselines and human-written skills across 86 tasks with up to 1000x cheaper inference.
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.
SSRP separates planning from execution in LLM agents to overcome the Attention Latch, delivering 715X resilience gains over ReAct baselines on MultiWOZ tasks.
Reshaping outcome rewards, process signals, and rollout comparability in GRPO raises strict compile-and-semantic accuracy in agentic code repair from 0.385 to 0.535 under weak feedback.
Agentic AI systems are shifting software engineering from line-level code generation to delegated repository-scale execution under supervision, with SWE-bench performance rising from 1.96% to 78.4% and productivity gains of 13.6-55.8%.
MDIA, a specialty-routed 7-node multi-agent system, reports 0.6272 accuracy on 525 HealthBench Professional cases using GPT-5.4, outperforming the ChatGPT for Clinicians baseline by 3.72 points and attributing the lift to architectural components.
citing papers explorer
-
SciIntegrity-Bench: A Benchmark for Evaluating Academic Integrity in AI Scientist Systems
SciIntegrity-Bench shows seven LLMs exhibit a 34.2% integrity failure rate in dilemmatic scenarios, with all models fabricating synthetic data in missing-data cases and an intrinsic completion bias persisting after prompt changes.
-
ContextEcho: A Benchmark for Persona Drift in Long Agentic-Coding Sessions
ContextEcho benchmark shows persona drift occurs across 23 frontier models in long agentic-coding sessions, is not reliably reset by compaction, and can be restored by single-shot anchors with mode-dependent effects.
-
MemGym: a Long-Horizon Memory Environment for LLM Agents
MemGym unifies agent gyms into a memory benchmark with isolated scoring across tool-use, research, coding, and computer-use regimes plus a lightweight reward model for tractable coding evaluation.
-
AI Harness Engineering: A Runtime Substrate for Foundation-Model Software Agents
The paper defines AI Harness Engineering as a runtime substrate with eleven components and a four-level ladder that reframes agent reliability as a model-harness-environment system property rather than model capability alone.
-
Predictive Maps of Multi-Agent Reasoning: A Successor-Representation Spectrum for LLM Communication Topologies
Successor-representation spectra of row-stochastic communication operators predict perturbation robustness, consensus speed, and error accumulation in multi-agent LLM topologies, with condition number showing perfect empirical rank correlation.
-
BioMedArena: An Open-source Toolkit for Building and Evaluating Biomedical Deep Research Agents
BioMedArena supplies a standardized open toolkit with 166 biomedical benchmarks, 75 tools, 6 harnesses, and 6 context strategies that improve 12 backbones and surpass prior SOTA by 15.01 points on average across 8 benchmarks.
-
MemFlow: Intent-Driven Memory Orchestration for Small Language Model Agents
MemFlow routes queries by intent to tiered memory operations, nearly doubling accuracy of a 1.7B SLM on long-horizon benchmarks compared to full-context baselines.
-
TADI: Tool-Augmented Drilling Intelligence via Agentic LLM Orchestration over Heterogeneous Wellsite Data
TADI shows that domain-specialized tools orchestrated by an LLM over dual structured and semantic databases can convert heterogeneous wellsite data into evidence-grounded drilling intelligence, with tool design mattering more than model scale.
-
MolmoWeb: Open Visual Web Agent and Open Data for the Open Web
Open 4B and 8B visual web agents achieve state-of-the-art results on browser benchmarks by predicting actions from screenshots and instructions, outperforming similar open models and some closed larger-model agents, with full release of data and code planned.
-
WMF-AM: Probing LLM Working Memory via Depth-Parameterized Cumulative State Tracking
WMF-AM is a depth-parameterized benchmark that measures LLMs' cumulative state tracking ability without scratchpads, validated on 28 models across arithmetic and non-arithmetic tasks with ablations confirming the construct.
-
OpenJarvis: Personal AI, On Personal Devices
OpenJarvis decomposes personal AI into Intelligence, Engine, Agents, Tools & Memory, and Learning primitives and applies LLM-guided spec search to produce on-device configurations that reach within 3.2 pp of cloud baselines on average across eight tasks.
-
Hypergraph Enterprise Agentic Reasoner over Heterogeneous Business Systems
HEAR uses a stratified hypergraph ontology to orchestrate evidence-driven multi-hop reasoning over heterogeneous business systems, reaching 94.7% accuracy on supply-chain root-cause tasks with open-weight models.
-
Nautilus: From One Prompt to Plug-and-Play Robot Learning
A typed-contract harness with containerized 'chambers' and robotics-specific agent skills lets a coding LLM turn a single natural-language prompt into working reproduction, evaluation, and deployment workflows for robot learning.
-
Evidence Over Plans: Online Trajectory Verification for Skill Distillation
SPARK generates environment-verified trajectories to compute PDI, enabling posterior skill distillation that outperforms no-skill baselines and human-written skills across 86 tasks with up to 1000x cheaper inference.
-
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.
-
Beyond the Attention Stability Boundary: Agentic Self-Synthesizing Reasoning Protocols
SSRP separates planning from execution in LLM agents to overcome the Attention Latch, delivering 715X resilience gains over ReAct baselines on MultiWOZ tasks.
-
Signal Reshaping for GRPO in Weak-Feedback Agentic Code Repair
Reshaping outcome rewards, process signals, and rollout comparability in GRPO raises strict compile-and-semantic accuracy in agentic code repair from 0.385 to 0.535 under weak feedback.
-
Agentic AI in the Software Development Lifecycle: Architecture, Empirical Evidence, and the Reshaping of Software Engineering
Agentic AI systems are shifting software engineering from line-level code generation to delegated repository-scale execution under supervision, with SWE-bench performance rising from 1.96% to 78.4% and productivity gains of 13.6-55.8%.
-
MDIA: A Multi-Agent Diagnostic Intelligence Pipeline on HealthBench Professional
MDIA, a specialty-routed 7-node multi-agent system, reports 0.6272 accuracy on 525 HealthBench Professional cases using GPT-5.4, outperforming the ChatGPT for Clinicians baseline by 3.72 points and attributing the lift to architectural components.