Under consequence-invariant posterior training and sparsity of coordinated harm patterns, the training mass on dangerous guarded Predictors is bounded by C_bad times R_shell.
hub Mixed citations
Superintelligent agents pose catastrophic risks: Can scientist ai offer a safer path?
Mixed citation behavior. Most common role is background (50%).
15
Pith papers citing it
Background
50% of classified citations
hub tools
citation-role summary
background 3
other 2
method 1
citation-polarity summary
representative citing papers
SciTrace embeds cumulative safety deliberation and trajectory-aware verification into scientific agent pipelines, claiming SOTA safety gains and detection of 78.8% of compositional risks missed by single-step checks.
NEO is a probabilistic neural model that induces compositional programs as a learned Language of Thought from non-textual observations and executes them via a shared transition model to enable explanation-driven generalization.
Characterizes spurious correlation mechanisms in preference optimization via mean spurious bias and causal-spurious correlation leakage, demonstrates irreducible vulnerability to distribution shift, and introduces tie training as selective mitigation with validation on log-linear models and empirica
Neighbor-Consistency Belief (NCB) measures LLM belief robustness across conceptual neighborhoods, revealing that high-NCB facts resist contextual interference better, and Structure-Aware Training reduces brittleness by about 30%.
A survey of 136 U.S. clinicians finds that autonomous AI prescribing would require confidence-based escalation, differentiated uncertainty communication, and inferential transparency to gain acceptance and properly allocate liability.
Agentic misalignment in multi-agent systems arises from generic utilities causing posterior collapse; Agentic Evidence Attribution using self-reflection or weak-to-strong generalization provides context-specific evidence to align agent posteriors.
AI value alignment is reconceptualized as a pluralistic governance problem arising along three axes—objectives, information, and principals—making it inherently context-dependent and unsolvable by technical design alone.
ADAPT is an online reweighting framework for LLM training that outperforms offline data selection and mixing methods in cross-benchmark generalization under equal compute.
LLM agents use a Cartesian split between learned prediction and engineered control, enabling modularity but creating sensitivity and bottlenecks unlike integrated biological systems.
AI risks arise from growth-oriented economies, and post-growth concepts such as satisficing, the Doughnut model, and resource caps can reduce those risks while prioritizing tool-like AI over agentic systems.
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 a four-role framework for LLMs in scientific innovation and reviews methods, benchmarks, and limitations across Assistant, Collaborator, Scientist, and Evaluator roles.
This survey frames foundation agents using brain-inspired modular architectures and reviews challenges in evolution, collaboration, and safety.
citing papers explorer
-
Safety from Honesty in a Disinterested AI Predictor
Under consequence-invariant posterior training and sparsity of coordinated harm patterns, the training mass on dangerous guarded Predictors is bounded by C_bad times R_shell.
-
SciTrace: Trajectory-Aware Safety Reasoning for Scientific Discovery Agents
SciTrace embeds cumulative safety deliberation and trajectory-aware verification into scientific agent pipelines, claiming SOTA safety gains and detection of 78.8% of compositional risks missed by single-step checks.
-
Learning to Theorize the World from Observation
NEO is a probabilistic neural model that induces compositional programs as a learned Language of Thought from non-textual observations and executes them via a shared transition model to enable explanation-driven generalization.
-
Spurious Correlation Learning in Preference Optimization: Mechanisms, Consequences, and Mitigation via Tie Training
Characterizes spurious correlation mechanisms in preference optimization via mean spurious bias and causal-spurious correlation leakage, demonstrates irreducible vulnerability to distribution shift, and introduces tie training as selective mitigation with validation on log-linear models and empirica
-
Illusions of Confidence? Diagnosing LLM Truthfulness via Neighborhood Consistency
Neighbor-Consistency Belief (NCB) measures LLM belief robustness across conceptual neighborhoods, revealing that high-NCB facts resist contextual interference better, and Structure-Aware Training reduces brittleness by about 30%.
-
The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
A survey of 136 U.S. clinicians finds that autonomous AI prescribing would require confidence-based escalation, differentiated uncertainty communication, and inferential transparency to gain acceptance and properly allocate liability.
-
A Sober Look at Agentic Misalignment in Automated Workflows
Agentic misalignment in multi-agent systems arises from generic utilities causing posterior collapse; Agentic Evidence Attribution using self-reflection or weak-to-strong generalization provides context-specific evidence to align agent posteriors.
-
Relative Principals, Pluralistic Alignment, and the Structural Value Alignment Problem
AI value alignment is reconceptualized as a pluralistic governance problem arising along three axes—objectives, information, and principals—making it inherently context-dependent and unsolvable by technical design alone.
-
Rethinking Data Curation in LLM Training: Online Reweighting Offers Better Generalization than Offline Methods
ADAPT is an online reweighting framework for LLM training that outperforms offline data selection and mixing methods in cross-benchmark generalization under equal compute.
-
The Cartesian Cut in Agentic AI
LLM agents use a Cartesian split between learned prediction and engineered control, enabling modularity but creating sensitivity and bottlenecks unlike integrated biological systems.
-
The economic alignment problem of artificial intelligence
AI risks arise from growth-oriented economies, and post-growth concepts such as satisficing, the Doughnut model, and resource caps can reduce those risks while prioritizing tool-like AI over agentic systems.
-
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.
-
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.
-
Advances and Challenges in Foundation Agents: From Brain-Inspired Intelligence to Evolutionary, Collaborative, and Safe Systems
This survey frames foundation agents using brain-inspired modular architectures and reviews challenges in evolution, collaboration, and safety.
- Unbiased Canonical Set-Valued Oracles Via Lattice Theory