A pre-registered experiment on Claude Opus 4.8 agents confirms an information-theoretic capacity region for coupled LLM-agent markets (with one key result being an algebraic identity) and finds that LLM populations exhibit attractor dynamics with dispersion collapse rather than smooth mean-field响应.
Title resolution pending
7 Pith papers cite this work. Polarity classification is still indexing.
7
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
years
2026 7roles
background 1polarities
background 1representative citing papers
A four-axis taxonomy and inheritance map for seven event-linked perpetual contract variants, with design constraints and evaluability criteria but no empirical estimates.
LLM reliability techniques are unified as communication channel operators, with a new cost-aware router achieving superior quality-cost tradeoffs on hard tasks.
Training a Gaussian last layer by direct minimization of a shared-cavity predictive log-loss yields single-pass NLL/calibration parity-or-better with tuned and ensembled baselines on 7/8 UCI and 4/5 large/deep benchmarks.
Under cached oracle responses, identification and output-quality ceilings are governed by total variation distance and expected transcript-computable score; fresh responses raise these ceilings at the Chernoff rate for identification and via geometric stopping for output quality.
In untrusted strategic settings with unreliable measurements, the quantum primitive for output-hiding function sharing permits parties to generate private unbiased coins.
Proposes a quantum primitive for output-hiding function sharing with applications to enhanced QKD security and hidden joint function encoding.
citing papers explorer
-
Information Limits and Attractor Dynamics in Economies of Frontier LLM Agents: A Pre-Registered Test
A pre-registered experiment on Claude Opus 4.8 agents confirms an information-theoretic capacity region for coupled LLM-agent markets (with one key result being an algebraic identity) and finds that LLM populations exhibit attractor dynamics with dispersion collapse rather than smooth mean-field响应.
-
A Taxonomy of Event-Linked Perpetual Futures: Design Axes, Failure Modes, and Empirical Evaluability
A four-axis taxonomy and inheritance map for seven event-linked perpetual contract variants, with design constraints and evaluability criteria but no empirical estimates.
-
A Communication-Theoretic Framework for LLM Agents: Cost-Aware Adaptive Reliability
LLM reliability techniques are unified as communication channel operators, with a new cost-aware router achieving superior quality-cost tradeoffs on hard tasks.
-
Direct Bethe Free Energy Minimization for Bayesian Neural Networks
Training a Gaussian last layer by direct minimization of a shared-cavity predictive log-loss yields single-pass NLL/calibration parity-or-better with tuned and ensembled baselines on 7/8 UCI and 4/5 large/deep benchmarks.
-
Computing with Stochastic Oracles in AI-Augmented Computation
Under cached oracle responses, identification and output-quality ceilings are governed by total variation distance and expected transcript-computable score; fresh responses raise these ceilings at the Chernoff rate for identification and via geometric stopping for output quality.
-
Quantum Primitive for Output-Hiding Function Sharing: Strategic Settings
In untrusted strategic settings with unreliable measurements, the quantum primitive for output-hiding function sharing permits parties to generate private unbiased coins.
-
Quantum Primitive for Output-Hiding Function Sharing: QKD and Joint Computation Applications
Proposes a quantum primitive for output-hiding function sharing with applications to enhanced QKD security and hidden joint function encoding.