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
AI4Research
Canonical reference. 91% of citing Pith papers cite this work as background.
23
Pith papers citing it
Background
91% of classified citations
hub tools
citation-role summary
background 11
citation-polarity summary
roles
background 11representative citing papers
DataPRM is an environment-aware generative process reward model that improves LLM data analysis agents by 7-11% on benchmarks via active verification and reflection-aware ternary rewards.
A survey of LLM copyright protection that unifies text watermarking, model watermarking, and model fingerprinting while presenting new coverage of fingerprint transfer and removal.
Formalizes judgment-grounded expansion as a human-AI collaborative task for peer review generation, supported by a user study and conformal prediction methods for scalable evaluation.
Introduces BAG, an open-source benchmark suite with baselines, datasets, and tasks for assessing Bayesian low-rank adaptation of multi-modal language models on calibration, robustness, and decision-making under uncertainty.
LLM-AutoSciLab proposes an LLM-driven closed-loop system for hypothesis generation and adaptive experiment selection that reports higher accuracy and 2-5x better sample efficiency than baselines on new chemistry and gene-network discovery benchmarks.
Empirical analysis across 15 LLMs and 1,141 skills identifies a logarithmic routing decay law and a multiplicative execution law coupled by a single fitted slope parameter b that enables targeted library optimizations improving routing accuracy and downstream task pass rates.
AI peer review systems are vulnerable to prompt injections, prestige biases, assertion strength effects, and contextual poisoning, as demonstrated by a new attack taxonomy and causal experiments on real conference submissions.
A think-aloud study reveals that AI tools in early research misrepresent uncertainty, obscure provenance, and create fragile trust, leading researchers to develop compensatory strategies to preserve scholarly judgment.
OMIBench benchmark reveals that current LVLMs achieve at most 50% on Olympiad problems requiring reasoning across multiple images.
MeasHalu reduces LLM hallucinations in scientific measurement extraction via a fine-grained taxonomy, reasoning-aware fine-tuning, and progressive rewards, improving accuracy on the MeasEval benchmark.
An empirical study on 20 architecture papers finds AI reviews capture a significant fraction of human-raised issues while also surfacing additional ones, using a released tool that clusters AI comments for comparison.
Eywa enables language-based agentic AI systems to collaborate with specialized scientific foundation models for improved performance on structured data tasks.
AblateCell reproduces baselines in three single-cell perturbation repositories with 88.9% success and recovers ground-truth critical components with 93.3% accuracy via closed-loop ablation.
Generative multi-agent systems exhibit emergent collusion and conformity behaviors that cannot be prevented by existing agent-level safeguards.
The paper unifies perspectives on Long CoT in reasoning LLMs by introducing a taxonomy, detailing characteristics of deep reasoning and reflection, and discussing emergence phenomena and future directions.
The paper proposes the Cybersecurity AI Scientist as a modular multi-agent architecture for automating cybersecurity research, distinguished by its focus on non-stationary threats and anchored in a four-zeros risk-trust-incident-energy frame.
A survey organizing AI-powered research automation into five workflow stages, defining AutoResearch and Vibe Research, and proposing five evaluation dimensions while noting domain-conditioned limits on autonomy.
SciAtlas builds a large-scale multi-disciplinary academic knowledge graph and a neuro-symbolic retrieval system to support automated scientific research tasks such as literature review and idea positioning.
AI can generate research artifacts faster than it can verify them, so across all eight lifecycle stages the credible deployment mode is human-governed collaboration rather than full autonomy.
SciHorizon-DataEVA is a multi-agent system that applies Sci-TQA2 principles across four dimensions to assess AI-readiness of heterogeneous scientific data via dynamic profiling and self-correcting evaluation workflows.
The survey organizes Context Engineering into retrieval, processing, management, and integrated systems like RAG and multi-agent setups while identifying an asymmetry where LLMs handle complex inputs well but struggle with equally sophisticated long outputs.
The paper proposes a four-role framework for LLMs in scientific innovation and reviews methods, benchmarks, and limitations across Assistant, Collaborator, Scientist, and Evaluator roles.
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.
-
Rewarding the Scientific Process: Process-Level Reward Modeling for Agentic Data Analysis
DataPRM is an environment-aware generative process reward model that improves LLM data analysis agents by 7-11% on benchmarks via active verification and reflection-aware ternary rewards.
-
Copyright Protection for Large Language Models: A Survey of Methods, Challenges, and Trends
A survey of LLM copyright protection that unifies text watermarking, model watermarking, and model fingerprinting while presenting new coverage of fingerprint transfer and removal.
-
Judgment-Grounded Expansion for Peer Review Generation
Formalizes judgment-grounded expansion as a human-AI collaborative task for peer review generation, supported by a user study and conformal prediction methods for scalable evaluation.
-
Bayesian Adaptation Gym: A Benchmark for the Bayesian Low-Rank Adaptation of Multi-Modal Language Models
Introduces BAG, an open-source benchmark suite with baselines, datasets, and tasks for assessing Bayesian low-rank adaptation of multi-modal language models on calibration, robustness, and decision-making under uncertainty.
-
LLM-AutoSciLab: Closed-Loop Scientific Discovery via Active Experimentation with LLMs
LLM-AutoSciLab proposes an LLM-driven closed-loop system for hypothesis generation and adaptive experiment selection that reports higher accuracy and 2-5x better sample efficiency than baselines on new chemistry and gene-network discovery benchmarks.
-
The Scaling Laws of Skills in LLM Agent Systems
Empirical analysis across 15 LLMs and 1,141 skills identifies a logarithmic routing decay law and a multiplicative execution law coupled by a single fitted slope parameter b that enables targeted library optimizations improving routing accuracy and downstream task pass rates.
-
When AI reviews science: Can we trust the referee?
AI peer review systems are vulnerable to prompt injections, prestige biases, assertion strength effects, and contextual poisoning, as demonstrated by a new attack taxonomy and causal experiments on real conference submissions.
-
How Researchers Navigate Accountability, Transparency, and Trust When Using AI Tools in Early-Stage Research: A Think-Aloud Study
A think-aloud study reveals that AI tools in early research misrepresent uncertainty, obscure provenance, and create fragile trust, leading researchers to develop compensatory strategies to preserve scholarly judgment.
-
OMIBench: Benchmarking Olympiad-Level Multi-Image Reasoning in Large Vision-Language Model
OMIBench benchmark reveals that current LVLMs achieve at most 50% on Olympiad problems requiring reasoning across multiple images.
-
MeasHalu: Mitigation of Scientific Measurement Hallucinations for Large Language Models with Enhanced Reasoning
MeasHalu reduces LLM hallucinations in scientific measurement extraction via a fine-grained taxonomy, reasoning-aware fine-tuning, and progressive rewards, improving accuracy on the MeasEval benchmark.
-
Can AI Review Improve Paper Drafting? An Empirical Study on 20 Computer Architecture Submissions
An empirical study on 20 architecture papers finds AI reviews capture a significant fraction of human-raised issues while also surfacing additional ones, using a released tool that clusters AI comments for comparison.
-
Heterogeneous Scientific Foundation Model Collaboration
Eywa enables language-based agentic AI systems to collaborate with specialized scientific foundation models for improved performance on structured data tasks.
-
AblateCell: A Reproduce-then-Ablate Agent for Virtual Cell Repositories
AblateCell reproduces baselines in three single-cell perturbation repositories with 88.9% success and recovers ground-truth critical components with 93.3% accuracy via closed-loop ablation.
-
Emergent Social Intelligence Risks in Generative Multi-Agent Systems
Generative multi-agent systems exhibit emergent collusion and conformity behaviors that cannot be prevented by existing agent-level safeguards.
-
Towards Reasoning Era: A Survey of Long Chain-of-Thought for Reasoning Large Language Models
The paper unifies perspectives on Long CoT in reasoning LLMs by introducing a taxonomy, detailing characteristics of deep reasoning and reflection, and discussing emergence phenomena and future directions.
-
Hephaestus: Toward a Cybersecurity AI Scientist
The paper proposes the Cybersecurity AI Scientist as a modular multi-agent architecture for automating cybersecurity research, distinguished by its focus on non-stationary threats and anchored in a four-zeros risk-trust-incident-energy frame.
-
AutoResearch AI: Towards AI-Powered Research Automation for Scientific Discovery
A survey organizing AI-powered research automation into five workflow stages, defining AutoResearch and Vibe Research, and proposing five evaluation dimensions while noting domain-conditioned limits on autonomy.
-
SciAtlas: A Large-Scale Knowledge Graph for Automated Scientific Research
SciAtlas builds a large-scale multi-disciplinary academic knowledge graph and a neuro-symbolic retrieval system to support automated scientific research tasks such as literature review and idea positioning.
-
AI for Auto-Research: Roadmap & User Guide
AI can generate research artifacts faster than it can verify them, so across all eight lifecycle stages the credible deployment mode is human-governed collaboration rather than full autonomy.
-
SciHorizon-DataEVA: An Agentic System for AI-Readiness Evaluation of Heterogeneous Scientific Data
SciHorizon-DataEVA is a multi-agent system that applies Sci-TQA2 principles across four dimensions to assess AI-readiness of heterogeneous scientific data via dynamic profiling and self-correcting evaluation workflows.
-
A Survey of Context Engineering for Large Language Models
The survey organizes Context Engineering into retrieval, processing, management, and integrated systems like RAG and multi-agent setups while identifying an asymmetry where LLMs handle complex inputs well but struggle with equally sophisticated long outputs.
-
Evolving Roles of LLMs in Scientific Innovation: Assistant, Collaborator, Scientist, and Evaluator
The paper proposes a four-role framework for LLMs in scientific innovation and reviews methods, benchmarks, and limitations across Assistant, Collaborator, Scientist, and Evaluator roles.