The Archive

LLMs generate optimized quantum circuits via evolutionary framework, automating discovery of efficient quantum algorithms under reversibility constraints.

Study reveals LLMs exhibit evaluation awareness in Indic languages, shifting behavior under perceived testing in non-English contexts.

Training method increases LLM diversity in document-grounded scientific ideation while maintaining quality across proposed research ideas.

Steering agent framework addresses goal drift in LLM-based multi-agent systems via dynamic intervention and context-aware course correction.

MindEval2 benchmark evaluates LLMs in mental health support using clinically grounded synthetic patients and physician-calibrated rubrics instead of opaque Likert scales.

Grid-Bench executable benchmark measures LLM ability to predict state changes across 2D-4D discrete worlds through sequential spatial operations.

AI Consilium multidimensional benchmark evaluates LLM clinical second opinions on 300 NEJM cases using physician-calibrated judge across eight model pipelines.

Subspace tuning during fine-tuning mitigates catastrophic forgetting in LLMs by constraining updates away from principal weight matrix directions.

CATCH applies continuous adversarial training to reduce dishonesty in LLMs while preserving capabilities on knowledge-grounded tasks.

Librarian agents proactively curate and reorganize long document corpora during inference-time preparation to anticipate diverse queries.

Policy gradient RL for LLM post-training loses reward invariance under regularized importance sampling; proposes baseline centering fix.

BitSparse losslessly compresses sparse ReLU activations during LLM training via custom GPU kernels, reducing memory footprint.

Mandarin LLM exhibits overuse of corrective construction 'not X, but Y'; proposes decision tree to distinguish substantive vs. defective uses.

Game-theoretic model reveals profit-maximizing safety investment in LLM supply chains falls below welfare-optimal and industry-optimal levels.

Framework evaluates LLM agents on integrated inquiry cycles: hypothesis generation, reasoning, and exploration in unified interactive setting.

Claude Opus 4.8 and Codex GPT Sol 5.6 fail at reverse-engineering iOS protobuf schemas with high error rates; LLMs unreliable for forensic analysis.

Empirical evaluation finds tool graph memory rarely improves multi-turn agent tool-use over nine models, with gains costing 2x additional LLM calls.

Hardware verification framework uses LLM agents with independent data-integrity validation via scoreboards for coverage closure in RTL designs.

BiShield-TEE identifies vulnerability in unilateral weight obfuscation for on-device LLMs via Solution Determinacy Property, exposing model structure.

Benchmark comparing Muon optimizer vs AdamW across data/time/tuning budgets on 17.7M–33.6M LLMs and vision transformers, showing 3.25–22.44% validation-loss gains.

Survey of RL-driven diffusion models for language/vision generation, examining reward-based trajectory optimization analogous to RLHF for LLMs.

Proposes Function-isolated Memory-Efficient Optimization (CiME) localizing low-overlap functional subnetworks in LLMs for continual learning without catastrophic forgetting.

Framework for using LLMs as game masters in tabletop RPGs, addressing story generation, rule adherence, and player action freedom.

SCRIBE framework produces neuroscience-grounded interpretable semantic representations from LLM ensembles; validates context-dependent human semantic judgments with 921 participants.

Poison Valleys diagnostic traces path-dependent failures in LLM code repair agents via hypothesis/implementation tracking on 1,844 requests across PersistQueue and Dijkstra tasks.

Variance partitioning decomposes LLaMA2-7B brain alignment using SCRIBE's 65 neurocognitive semantic properties on Moth Radio Hour fMRI narratives.

ISW-Agent integrates LLM with n8n workflows for internal solitary wave remote sensing; achieves 94.2% task-match rate on 120 Chinese instructions.

MetaExp enables LLM agents to learn transferable experience across tasks via scalable cross-task environment feedback.

Tool-augmented LLM framework separates variable extraction, formula execution, and formatting for reliable medical calculator scoring from patient notes.

Benchmark of 83 published works shows LLMs struggle to generate nonlinear narratives, preferring chronological event ordering over author-specific structures.