Outputs using selected claims had more support, fewer hallucinations
Preprint: Outputs using selected English claims had more supported additions and fewer hallucinations on average, but results varied by language and generator.
Search
2,451 results
1741–1760
Preprint finds trading order can shape outcomes in a fee-free market
A mathematical two-trader model links interior equilibrium to price matching and finds first-mover effects, while reserve scale changes numerical welfare comparisons.
AI can tidy model cards, but generated errors persist
Preprint: Two LLM workflows kept most existing information in place, but only 26 of 48 newly generated cards were judged fully correct.
AI retrieval system scores higher in crystal route planning
Preprint: A retrieval-based language model outperformed an LLM-only baseline and was used to plan routes for prospective target phases.
GeoFormer posts lowest error in four-site seismic picking test
A preprint reports lower first-arrival picking RMSE than a Vision Transformer on Brunswick, Halfmile and Dongbei, while its noise evidence remains qualitative.
Study Reports Higher Bridge-Damage Scores After Image Restoration
Preprint: The study reports higher scores when a detector-agnostic restoration model processes synthetic pairs and 200 real nighttime UAV images, while real-world validation remains limited.
RF Simulation Finds Learned Labels Hold Together in Heavy Noise
Preprint: A controlled 3-D simulation found semantic labels remained coherent deeper into noise, while fading-input tests mislabelled 18% of held-out novel objects as known.
Robot tests favor geometric targets over text-generated coordinates
This arXiv Preprint compares geometry-specific readouts with text-coordinate output across grounding benchmarks, simulated manipulation, transfer tests and physical robot trials.
New method maps how atomic motion changes magnetic exchange
A preprint reports a two-orders-of-magnitude cost advantage in SrMnO3, but the method remains a limited computational demonstration.
Classical Theory Reproduces Radiation Relations, Not Blackbody Spectrum
Preprint: A theoretical analysis finds that classical fields recover links between radiation energy, pressure and flux, but not the spectrum’s absolute normalization.
Flight-control review finds trade-offs among NDI architectures
Preprint: A review compares model-based, sensory, incremental and hybrid nonlinear dynamic inversion, including a Citation II flight-test example.
Mixed training scores higher than next-chunk reasoning in AI tests
An arXiv preprint reports higher post-RLVR scores and sharply lower compute costs for the mixed training route.
Laser study maps how ionized D2O forms a D-D bond
Preprint: Measurements and simulations follow D2+ formation through an indirect pathway with direct, roaming and delayed branches.
DF-MoE tops MAVOS-DD tests, but wider scores vary
An arXiv preprint reports the highest displayed scores across four MAVOS-DD scenarios, while AVLips, cross-dataset and video-only comparisons are less uniform.
New statistical framework adds an inconclusive option to research
A statistical preprint proposes a way to classify estimates into rival possibilities while controlling the risk of assigning the wrong class.
ECG adaptation method reports gains across three test settings
Preprint: The method was tested on three ECG datasets under offline, continual online and independent online adaptation protocols.
Arabic NLP research surges, but dialect coverage remains uneven
An arXiv version 2 preprint dated 25 August 2026 maps 7,120 papers and finds core-task coverage centered on Modern Standard Arabic.
Preprint reports higher benchmark scores for a clinical AI agent
MediSkill-Evo led several fixed benchmark comparisons, but the results do not establish clinical safety or prospective validity.
Small AI Models Transfer Across Domains Without Local Training Text
This preprint reports lower judged error rates for compact models, but its benchmark and language-model evaluation limit the result.
Multimodal training controller leads on two benchmarks but trails on VGGSound
An arXiv version 1 preprint reports strong results on CREMA-D and AV-MNIST, mixed results elsewhere, and no win over VGGSound’s strongest baselines.