Author-Reviewed Papers

For every paper on this page, at least one of the original authors has seen our plain-language explanation and engaged with it — either confirming it reads accurately or requesting corrections that we then applied. An endorsement does not mean the authors formally approve every sentence, but it does mean the explanation has passed the eyes of the people who wrote the paper.

1351 papers reviewed by authors · 801–810 / 1351

🔭 astrophysics

Learning the Universe with the 2nd Generation of CAMELS: Varying 35 parameters of the IllustrisTNG model in (50Mpc/h)^3 boxes

This paper presents the second generation of the CAMELS project, featuring 1,192 cosmological simulations with eight times larger volumes than its predecessor to explore a 35-parameter space, demonstrating that while these larger volumes improve neural network-based parameter inference, the gains are sub-linear due to mode coupling and parameter degeneracies.

Shy Genel, Yongseok Jo, Boon Kiat Oh, Megan Taylor Tillman, Max E. Lee, Jun-Young Lee, Elena Hernández-Martínez, Christo (…)2026-06-10✓ Author reviewed ⓘ

🤖 AI

An LLM-Native Psychometric Instrument Does Not Predict LLM Behavior: Evidence Across 25 Models

Despite demonstrating high internal reliability, a novel psychometric instrument derived from LLM behavioral affordances fails to predict actual LLM behavior, revealing a critical disconnect between model self-reports and observed actions that poses a significant risk for LLM-as-judge evaluation pipelines.

Juan Manuel Contreras2026-06-10✓ Author reviewed ⓘ

⚛️ quantum physics

Most incompatible measurements and sum-of-squares optimisation

This paper advances the quantification of measurement incompatibility in finite-dimensional quantum systems by deriving analytical universal bounds through novel parent measurements, formalizing their construction via sum-of-squares optimization, and demonstrating their application in certifying high-dimensional quantum steering.

Sébastien Designolle2026-06-10✓ Author reviewed ⓘ

💬 NLP

$\tau$ -Rec: A Verifiable Benchmark for Agentic Recommender Systems

The paper introduces $\tau$ -Rec, a verifiable benchmark for agentic recommender systems that replaces subjective LLM-based evaluations with structured rewards and a reveal-tagged elicitation mechanism, revealing a significant reliability gap in current conversational agents where even top models struggle to consistently meet task constraints.

Bharath Sivaram Narasimhan, Karthik R Narasimhan2026-06-10✓ Author reviewed ⓘ

🔬 condensed matter

Dissipative response of driven bead-spring-dashpot chains

This paper numerically demonstrates that while the dissipated work in pulling a polymer chain without internal friction consistently increases with chain length, the presence of internal friction introduces a stiffness-dependent relationship where dissipation either increases or decreases with chain length depending on the pulling trap stiffness, thereby invalidating the simple damping-dissipation correlation observed in single-mode systems.

R. Kailasham2026-06-10✓ Author reviewed ⓘ

🧬 biology

Chemotaxing E. coli do not count single molecules

This study demonstrates that *E. coli* chemotaxis is limited by internal noise in signal processing rather than the physical limits of molecule diffusion, as the bacteria encode two orders of magnitude less information than theoretically possible.

Henry H. Mattingly, Keita Kamino, Jude Ong, Rafaela Kottou, Thierry Emonet, Benjamin B. Machta2026-06-10✓ Author reviewed ⓘ

💻 computer science

Experimental evaluation of optimal abstract operators for sharing and linearity analysis

This paper experimentally evaluates the trade-off between precision and performance in static analysis of logic programs by implementing and testing optimal abstract operators for sharing and linearity analysis within the CiaoPP preprocessor.

Francesca Scozzari, Gianluca Amato2026-06-10✓ Author reviewed ⓘ

💻 computer science

On the Complexity of the Bi-infinite Post Correspondence Problem

This paper establishes that the bi-infinite Post Correspondence Problem ( $\mathbb{Z}$ PCP) is $\Sigma^0_2$ -complete within the arithmetical hierarchy by presenting a chain of reductions from Turing machine non-halting and proving the $\Pi^0_1$ -completeness of several related infinite and shifted variants.

Olivier Finkel, Vesa Halava2026-06-10✓ Author reviewed ⓘ

💻 computer science

ClawXiv: a signed archival workflow and distributed publication architecture for human--AI collaborative research

This paper introduces ClawXiv, a local, author-side workflow and distributed architecture designed to transform volatile human-AI collaborative research processes into durable, signed, and inspectable archival artifacts through a four-state migration pipeline and a suite of operational scripts.

Andras Kornai2026-06-09✓ Author reviewed ⓘ

📊 statistics

On the statistical analysis of grouped data: when Pearson $χ^2$ and other divisible statistics are not goodness-of-fit tests

This paper challenges the common assumption that divisible statistics like Pearson's $\chi^2$ serve as effective goodness-of-fit tests in sparse data regimes with many bins, proposing instead a unifying framework that reveals the limitations of existing methods and offers modified, more powerful alternatives along with new distribution-free tests.

Sara Algeri, Estate V. Khmaladze2026-06-09✓ Author reviewed ⓘ

← Previous Next →

Learning the Universe with the 2nd Generation of CAMELS: Varying 35 parameters of the IllustrisTNG model in (50Mpc/h)^3 boxes

An LLM-Native Psychometric Instrument Does Not Predict LLM Behavior: Evidence Across 25 Models

Most incompatible measurements and sum-of-squares optimisation

τ\tauτ-Rec: A Verifiable Benchmark for Agentic Recommender Systems

Dissipative response of driven bead-spring-dashpot chains

Chemotaxing E. coli do not count single molecules

Experimental evaluation of optimal abstract operators for sharing and linearity analysis

On the Complexity of the Bi-infinite Post Correspondence Problem

ClawXiv: a signed archival workflow and distributed publication architecture for human--AI collaborative research

On the statistical analysis of grouped data: when Pearson χ2χ^2χ2 and other divisible statistics are not goodness-of-fit tests

$\tau$ -Rec: A Verifiable Benchmark for Agentic Recommender Systems

On the statistical analysis of grouped data: when Pearson $χ^2$ and other divisible statistics are not goodness-of-fit tests