Rule2DRC is a benchmark for LLM agents synthesizing DRC scripts from natural language rules, paired with SplitTester that improves Best-of-N selection via execution-guided discriminative test generation.
hub
Advances in neural information processing systems , volume=
27 Pith papers cite this work. Polarity classification is still indexing.
27
Pith papers citing it
hub tools
citation-role summary
background 1
dataset 1
citation-polarity summary
years
2026 27representative citing papers
ClawForge is a generator framework that creates reproducible executable benchmarks for command-line agents under state conflict, with ClawForge-Bench showing frontier models reach at most 45.3% strict accuracy and that state inspection drives most performance gaps.
Semantic consensus on model outputs for public prompts enables federated LLM fine-tuning that matches parameter-aggregation baselines with orders-of-magnitude lower communication.
DOLORES, an agent using a formal language for meta-reasoning to construct adaptive scaffolds on the fly, outperforms prior scaffolding methods by 24.8% on average across four hard benchmarks and multiple model sizes.
A new RL Feasibility Index based on task learnability via reinforcement learning diverges from prior AI exposure measures, rating operational jobs like power plant operators as highly feasible while rating creative and interpersonal roles as less so.
MasterSet is a new large-scale benchmark for must-cite citation recommendation in AI/ML, using LLM-annotated tiers on 150k papers and Recall@K evaluation.
Validity-calibrated reasoning distillation improves transfer of reasoning skills by modulating updates based on relative local validity of next steps instead of enforcing full trajectory imitation.
Prediction agreement between open and closed LLMs substantially overstates agreement on attributions and causal reasons.
SWE-Router introduces trajectory-conditioned value-based routing for LLM agents on SWE tasks, with a Bayes-optimality theorem and empirical cost savings while retaining most strong-model performance.
TheraJudge, trained via preference optimization on human annotations, reaches high clinician agreement (ICC 0.87-0.95) and, when used by TheraAgent, raises human-rated therapeutic quality by 0.43 points on a 5-point scale with 94% recovery of low-quality responses.
COALA applies convex optimization reformulations of neural networks to direct preference optimization, claiming single-GPU training with ~18% of DPO's TFLOPs and competitive performance on multiple datasets and models up to 8B parameters.
BalanceRAG uses sequential graphical testing on a 2D lattice of threshold pairs to certify safe operating points that meet target risk levels in cascaded RAG while increasing coverage.
Introduces the TCR framework to evaluate educational LLM assistants on transparency, consistency, and refinement in multi-turn interactions, complementing aggregate metrics.
A hierarchical genetic algorithm induces overthinking in black-box large reasoning models by perturbing logical structure, achieving up to 26.1x longer outputs on the MATH benchmark.
Attention to goal tokens declines in multi-turn LLM interactions while residual representations often retain decodable goal information, and the gap between these predicts whether goal-conditioned behavior survives.
LegalCiteBench reveals that current LLMs achieve under 7% accuracy on closed-book legal citation retrieval and completion tasks, with misleading answer rates above 94% for nearly all tested models.
HAGE proposes a trainable weighted graph memory framework with LLM intent classification, dynamic edge modulation, and RL optimization that improves long-horizon reasoning accuracy in agentic LLMs over static baselines.
Language models engage in counterexample-repair loops for conceptual definitions but produce increasingly verbose outputs without accuracy gains and hit diminishing returns quickly.
MultiBreak is a large diverse multi-turn jailbreak benchmark that achieves substantially higher attack success rates on LLMs than prior datasets and reveals topic-specific vulnerabilities in multi-turn settings.
Small LLMs under 2B parameters achieve better economic break-even, energy efficiency, and hardware density than larger models on legacy GPUs for industrial tasks.
Orthonormal initialization for LoRA in RLVR achieves the minimal gap to full fine-tuning, stabilizes training, and outperforms standard LoRA and prior variants on mathematical reasoning benchmarks.
LLMs exhibit reproducible asymmetries in advice on hypothetical religious conversions, favoring Catholic, Bahá'í, and Sikh transitions while disfavoring Atheist, Agnostic, and Jehovah's Witness ones across 20 models and 182 pairings.
CR4T is a model-agnostic framework using lightweight risk detection and domain-conditioned rewriting to convert unsafe or refusal-style LLM responses into developmentally appropriate guidance for adolescents.
Learning a margin-based confidence ranker for LLM judges improves agreement-target success in cascaded selective evaluation compared to heuristic confidence scores.
citing papers explorer
-
Rule2DRC: Benchmarking LLM Agents for DRC Script Synthesis with Execution-Guided Test Generation
Rule2DRC is a benchmark for LLM agents synthesizing DRC scripts from natural language rules, paired with SplitTester that improves Best-of-N selection via execution-guided discriminative test generation.
-
ClawForge: Generating Executable Interactive Benchmarks for Command-Line Agents
ClawForge is a generator framework that creates reproducible executable benchmarks for command-line agents under state conflict, with ClawForge-Bench showing frontier models reach at most 45.3% strict accuracy and that state inspection drives most performance gaps.
-
Beyond Parameter Aggregation: Semantic Consensus for Federated Fine-Tuning of LLMs
Semantic consensus on model outputs for public prompts enables federated LLM fine-tuning that matches parameter-aggregation baselines with orders-of-magnitude lower communication.
-
Deep Reasoning in General Purpose Agents via Structured Meta-Cognition
DOLORES, an agent using a formal language for meta-reasoning to construct adaptive scaffolds on the fly, outperforms prior scaffolding methods by 24.8% on average across four hard benchmarks and multiple model sizes.
-
What Jobs Can AI Learn? Measuring Exposure by Reinforcement Learning
A new RL Feasibility Index based on task learnability via reinforcement learning diverges from prior AI exposure measures, rating operational jobs like power plant operators as highly feasible while rating creative and interpersonal roles as less so.
-
MasterSet: A Large-Scale Benchmark for Must-Cite Citation Recommendation in the AI/ML Literature
MasterSet is a new large-scale benchmark for must-cite citation recommendation in AI/ML, using LLM-annotated tiers on 150k papers and Recall@K evaluation.
-
Validity-Calibrated Reasoning Distillation
Validity-calibrated reasoning distillation improves transfer of reasoning skills by modulating updates based on relative local validity of next steps instead of enforcing full trajectory imitation.
-
Surrogate Fidelity: When Can Open LLMs Explain Closed Ones?
Prediction agreement between open and closed LLMs substantially overstates agreement on attributions and causal reasons.
-
SWE-Router: Routing in Multi-turn Agentic Software Engineering Tasks
SWE-Router introduces trajectory-conditioned value-based routing for LLM agents on SWE tasks, with a Bayes-optimality theorem and empirical cost savings while retaining most strong-model performance.
-
Training Therapeutic Judges and Multi-Agent Systems for Human-Aligned Mental Health Support
TheraJudge, trained via preference optimization on human annotations, reaches high clinician agreement (ICC 0.87-0.95) and, when used by TheraAgent, raises human-rated therapeutic quality by 0.43 points on a 5-point scale with 94% recovery of low-quality responses.
-
Convex Optimization for Alignment and Preference Learning on a Single GPU
COALA applies convex optimization reformulations of neural networks to direct preference optimization, claiming single-GPU training with ~18% of DPO's TFLOPs and competitive performance on multiple datasets and models up to 8B parameters.
-
BalanceRAG: Joint Risk Calibration for Cascaded Retrieval-Augmented Generation
BalanceRAG uses sequential graphical testing on a 2D lattice of threshold pairs to certify safe operating points that meet target risk levels in cascaded RAG while increasing coverage.
-
Evaluating Multi-turn Human-AI Interaction
Introduces the TCR framework to evaluate educational LLM assistants on transparency, consistency, and refinement in multi-turn interactions, complementing aggregate metrics.
-
Inducing Overthink: Hierarchical Genetic Algorithm-based DoS Attack on Black-Box Large Language Reasoning Models
A hierarchical genetic algorithm induces overthinking in black-box large reasoning models by perturbing logical structure, achieving up to 26.1x longer outputs on the MATH benchmark.
-
When Attention Closes: How LLMs Lose the Thread in Multi-Turn Interaction
Attention to goal tokens declines in multi-turn LLM interactions while residual representations often retain decodable goal information, and the gap between these predicts whether goal-conditioned behavior survives.
-
LegalCiteBench: Evaluating Citation Reliability in Legal Language Models
LegalCiteBench reveals that current LLMs achieve under 7% accuracy on closed-book legal citation retrieval and completion tasks, with misleading answer rates above 94% for nearly all tested models.
-
HAGE: Harnessing Agentic Memory via RL-Driven Weighted Graph Evolution
HAGE proposes a trainable weighted graph memory framework with LLM intent classification, dynamic edge modulation, and RL optimization that improves long-horizon reasoning accuracy in agentic LLMs over static baselines.
-
The Counterexample Game: Iterated Conceptual Analysis and Repair in Language Models
Language models engage in counterexample-repair loops for conceptual definitions but produce increasingly verbose outputs without accuracy gains and hit diminishing returns quickly.
-
MultiBreak: A Scalable and Diverse Multi-turn Jailbreak Benchmark for Evaluating LLM Safety
MultiBreak is a large diverse multi-turn jailbreak benchmark that achieves substantially higher attack success rates on LLMs than prior datasets and reveals topic-specific vulnerabilities in multi-turn settings.
-
Are Large Language Models Economically Viable for Industry Deployment?
Small LLMs under 2B parameters achieve better economic break-even, energy efficiency, and hardware density than larger models on legacy GPUs for industrial tasks.
-
Geometry-Preserving Orthonormal Initialization for Low-Rank Adaptation in RLVR
Orthonormal initialization for LoRA in RLVR achieves the minimal gap to full fine-tuning, stabilizes training, and outperforms standard LoRA and prior variants on mathematical reasoning benchmarks.
-
When AI Takes Sides on Questions of Faith: Persistent Asymmetries in AI-Mediated Faith Guidance
LLMs exhibit reproducible asymmetries in advice on hypothetical religious conversions, favoring Catholic, Bahá'í, and Sikh transitions while disfavoring Atheist, Agnostic, and Jehovah's Witness ones across 20 models and 182 pairings.
-
CR4T: Rewrite-Based Guardrails for Adolescent LLM Safety
CR4T is a model-agnostic framework using lightweight risk detection and domain-conditioned rewriting to convert unsafe or refusal-style LLM responses into developmentally appropriate guidance for adolescents.
-
Margin-Adaptive Confidence Ranking for Reliable LLM Judgement
Learning a margin-based confidence ranker for LLM judges improves agreement-target success in cascaded selective evaluation compared to heuristic confidence scores.
-
GiVA: Gradient-Informed Bases for Vector-Based Adaptation
GiVA uses gradients to initialize vector adapters so they match LoRA performance at eight times lower rank while keeping extreme parameter efficiency.
-
JD Oxygen AI Item Center (Oxygen AIIC) V1: An Industrial-Scale LLM/VLM-Centric Solution for Item Understanding, Management, and Applications
Oxygen AIIC is an industrial platform using LLMs and VLMs for scalable item knowledge production and service at JD.com, reporting 94.2% precision and 82.8% recall along with business metric improvements.
- To Call or Not to Call: A Framework to Assess and Optimize LLM Tool Calling