InfiniteScienceGym procedurally generates unbounded scientific repositories with exact ground-truth QA pairs to benchmark LLMs on data reasoning, abstention, and tool use without static datasets.
hub Canonical reference
Single-minus gluon tree amplitudes are nonzero
Canonical reference. 100% of citing Pith papers cite this work as background.
hub tools
citation-role summary
citation-polarity summary
years
2026 10roles
background 7polarities
background 7representative citing papers
Structured critic-actor loops improve AI performance on theoretical physics reasoning tasks, with benefits strongest in asymmetric model pairings using constructive feedback.
S-matrix consistency forces the complete gluon amplitude structure and requires Yang-Mills Lie algebra plus Higgs mechanism for unitarised massive vector boson scattering.
Small 7B reasoning models were fine-tuned on synthetic and curated QFT problems using RL and SFT, yielding performance gains, error analysis, and public release of data and traces.
k-server-bench formulates potential-function discovery for the k-server conjecture as a code-based inequality-satisfaction task; current agents fully solve the resolved k=3 case and reduce violations on the open k=4 case.
All self-dual theories with or without higher-spin fields possess nontrivial tree-level amplitudes in Kleinian or complex Minkowski kinematics, completing the celestial analogue of the higher-spin duality.
A Carrollian theory on null infinity reproduces all MHV and NMHV Yang-Mills tree amplitudes, with a new explicit NMHV expression.
Loop-level Carrollian amplitudes in N=4 SYM and N=8 supergravity are differential operators on tree-level versions, with logarithmic eikonal behavior and IR-safe factorization via natural splitting.
LLM agents produce outputs that meet basic functional criteria for creativity but lack the process-level, social, and personal elements required for ontological creativity.
AI agents exploring Platonic mathematical structures via proof hypergraphs may reveal the overall architecture of formal mathematics and what makes parts of it human-accessible.
citing papers explorer
-
InfiniteScienceGym: An Unbounded, Procedurally-Generated Benchmark for Scientific Analysis
InfiniteScienceGym procedurally generates unbounded scientific repositories with exact ground-truth QA pairs to benchmark LLMs on data reasoning, abstention, and tool use without static datasets.
-
When Does Critique Improve AI-Assisted Theoretical Physics? SCALAR: Structured Critic--Actor Loop for Agentic Reasoning
Structured critic-actor loops improve AI performance on theoretical physics reasoning tasks, with benefits strongest in asymmetric model pairings using constructive feedback.
-
Consistent Scattering Amplitudes, Yang-Mills, the Higgs Mechanism and the EFTs Beyond
S-matrix consistency forces the complete gluon amplitude structure and requires Yang-Mills Lie algebra plus Higgs mechanism for unitarised massive vector boson scattering.
-
Fine-Tuning Small Reasoning Models for Quantum Field Theory
Small 7B reasoning models were fine-tuned on synthetic and curated QFT problems using RL and SFT, yielding performance gains, error analysis, and public release of data and traces.
-
$k$-server-bench: Automating Potential Discovery for the $k$-Server Conjecture
k-server-bench formulates potential-function discovery for the k-server conjecture as a code-based inequality-satisfaction task; current agents fully solve the resolved k=3 case and reduce violations on the open k=4 case.
-
Amplitudes in self-dual (higher-spin) theories
All self-dual theories with or without higher-spin fields possess nontrivial tree-level amplitudes in Kleinian or complex Minkowski kinematics, completing the celestial analogue of the higher-spin duality.
-
Towards a Carrollian Description of Yang-Mills
A Carrollian theory on null infinity reproduces all MHV and NMHV Yang-Mills tree amplitudes, with a new explicit NMHV expression.
-
On Carrollian Loop Amplitudes for Gauge Theory and Gravity
Loop-level Carrollian amplitudes in N=4 SYM and N=8 supergravity are differential operators on tree-level versions, with logarithmic eikonal behavior and IR-safe factorization via natural splitting.
-
On the Creativity of AI Agents
LLM agents produce outputs that meet basic functional criteria for creativity but lack the process-level, social, and personal elements required for ontological creativity.
-
Artificial Intelligence and the Structure of Mathematics
AI agents exploring Platonic mathematical structures via proof hypergraphs may reveal the overall architecture of formal mathematics and what makes parts of it human-accessible.