Papers

Flash-dLLM optimizes KV caching and parallel decoding for diffusion LLMs via IO-aware techniques, addressing inference bottlenecks in non-autoregressive text generation.

Framework for decentralized multi-agent decision-making under partial observability with delayed information sharing using low-rank model learning.

Agensh scales multi-agent systems to 1,024 agents using decentralized self-organized coordination without central orchestrator bottleneck.

SpeakerMem-R1 introduces speaker-centered dual-track memory for multi-party dialogue, addressing person/group attribution and temporal state tracking.

CliffCompaction reduces token compression costs by 50% for long-horizon coding agents while maintaining performance on KernelBench and Terminal-Bench.

SWE-Serve benchmark evaluates agents on production inference engineering tasks spanning model support, runtime execution, and public APIs.

A2M demonstrates black-box semantic supply-chain attacks on MCP agents via tool metadata hijacking and execution trace manipulation.

Growing Harness learns reusable executable agent scaffolds from task feedback, reducing redundant LLM inference on repeated control decisions.

Study showing typed decision models may misinterpret option semantics despite schema conformance, demonstrating gap between syntax and intended semantics.

FleXray is a generalist X-ray segmentation model spanning whole-body anatomy for clinical analysis, addressing 2D projection ambiguity.

EquivSVA dataset enables testing whether LLM-generated SystemVerilog assertions capture true behavior vs. implementation-specific details.

Study shows compile rate is unreliable metric for LLM code vulnerability repair; proposes change-aware evaluation across 350M–6.7B parameter models.

A-DLCC proposes parameter-free clustering via β-integrated local depth; orthogonal to LLM/AI frontiers.

DISCO applies diffusion-based spatial attention to graph community detection; limited relevance to core AI systems.

RCT with 100 product professionals shows Figma Make's prompt-to-design tools reduce design task time; empirical productivity evidence.

LYRA identifies Proximity Trap: irrelevant nearby context overshadows distant evidence in long-context LLMs; proposes t-distributed relevance alignment.

TraceVIC uses causal reasoning over code evolution to identify vulnerability-inducing commits; improves on git-blame heuristics.

On-policy distillation (OPD) fixes quantization exposure bias in sub-3-bit LLMs; restores math and code reasoning in low-bit models.

Framework for optimal sequential annotation budgets in off-policy evaluation when LLM-as-judge labels carry unknown bias.

Theoretical physics study on learning-theoretic conditions for bosonic Gaussian states; unrelated to AI systems or LLMs.

Method for steering LLM reasoning via semantic exploration of problem-specific strategies rather than naive repeated sampling.

Serving infrastructure (Ollama, Gemma, Phi) confounds tool-use evaluation results; model behavior inconsistency stems from serving layer gatekeeping.

Stylometric classifier (Random Forest, ROC-AUC 0.87) detects ChatGPT-assisted student writing with 22% false positive rate.

PROSWIN: deep distributional regression forecasts hourly solar wind speed from solar images with 4-day lead time.

Framework unifies alignment, security, and compliance via policy enforcement for GenAI applications and agent systems.

Spectral theory explains grokking as transition from NTK regime to feature learning via weight decay-induced kernel evolution.

MAGIC: reinforcement learning generates task-specific multi-agent collaboration topologies with mixed granularity to reduce cost and improve performance.

Discovery-Driven Integration uses unstructured text to identify missing relational structure for joining semantically related tables in data lakes.

Parametric convergence rates for entropic optimal transport in semi-discrete regime with subGaussian measures and no dimension dependence.

Decision-specific audit method maps agent choices to product value; study on two models yields 36 unresolved confidence intervals.

GravityOCR uses diffusion-based parallel decoding with AR verification to accelerate document OCR inference beyond sequential token generation.

Method extracts hidden chain-of-thought reasoning from closed-source frontier models including GPT-6 Astra via API tool registration to validate reasoning quality.

Knowledge Pull Requests framework automates incremental document updates by extracting claims, routing to sections, and flagging content conflicts with change tracking.

Benchmark compares transfer learning, active learning, and pseudo-labeling for label-efficient cloud classification on Ground-based Cloud Dataset.

PersonaWeaver controls LLM-generated character diversity in procedural generation by mitigating behavioral homogeneity through structured persona attributes.

Information-theoretic criteria for selecting basis functions in sparse Gaussian process regression to optimize M-budget allocation across candidate bases.

Greedy LLM decoding produces different outputs across BF16/FP16 precision with 49-100% prompt divergence, challenging determinism assumptions in inference.

MMAP pretraining framework handles missing multimodal data and incomplete records for longitudinal Alzheimer's disease progression prediction.

Framework identifies semantic abstraction gaps in LLM natural language inference via constructed higher-order semantic knowledge and reasoning diagnostics.

Foundation model embeddings from screening mammograms encode pre-diagnostic tissue changes detectable before cancer diagnosis, varying by pretraining domain.

Hierarchical GNNs with latent communication improve power-flow model generalization across grid topologies using Kron-derived graph reductions.

Mixture-of-experts with diffusion models for cross-scenario physical layer security in 6G wireless networks.

GTR: softmax-free recurrent vision backbone using gated linear attention and spatial scans for efficient dense prediction, distilled from DINOv3.

Polyak-type step-size selection for extragradient methods solving monotone root-finding problems with unified deterministic analysis.

Language models conflate sycophancy with receptiveness; social psychology framework shows behavioral overlap creates construct-validity problems in current evals.

Quantum-aided active device detection for energy-harvesting symbiotic radio networks using NOMA in IoT deployments.

Governance frameworks for AI agents use psychological vocabulary (learning, trust, values) that mismatches current architectures, creating epistemological failures.

Neutral-atom quantum computing applied to NP-hard resource allocation problems (MAP) in NOMA wireless networks.

Decision-only LLM judge scales evaluation cost-effectively to 0.36% of full-judge fees, matches state-of-the-art on preference tasks, flags uncertain escalations.

UFO-MGen: flow-based generative model for crystal structure design learning Wyckoff topological representations for materials discovery.