Claims Browser

A causal recommendation architecture using existing holdback data can reduce recommendation impressions by 7% without reducing overall content consumption

"In a production-scale A/B test on millions of Spotify users, this policy reduces recommendation impressions by 7% with no statistically significant reduction in overall recommended content consumption."

Joint training with holdback data improves calibration of the treated head relative to production baseline

"joint training with holdback data improves calibration of the treated head relative to the production baseline"

Causal recommendation architectures can be built using existing experimentation infrastructure without requiring new data collection

"extending an existing production recommendation model to a causal architecture using holdback data that is already collected as part of routine experimentation infrastructure, requiring no new data collection"

A randomized study involving over 1,000 students has been conducted examining the relationship between ChatGPT usage and critical thinking, originality, and student performance on university assignments

"A randomized study of more than 1,000 students examines ChatGPT, critical thinking, originality, and student performance on a real-world university assignment."

AraMS-28k is the largest publicly released line-level dataset of genuine historical Arabic manuscripts with 28,600 annotated text lines

"We introduce AraMS-28k, the largest publicly released line-level dataset of genuine historical Arabic manuscripts, comprising 14 books, 3,043 pages, and 28,600 annotated text lines"

This is the first annotation released for a historical Arabic manuscript corpus that recovers non-linear reading order at line-level granularity

"margin lines that have an unambiguous attachment point in the main text are further annotated with an insertion anchor, recovering the manuscript's true non-linear reading order at line-level granularity -- to our knowledge the first such annotation released for a historical Arabic manuscript corpus"

RefLAM pipeline combines multimodal-LLM OCR with human review to construct high-quality manuscript datasets

"The dataset was constructed with RefLAM, a reference-grounded annotation pipeline that aligns multimodal-LLM OCR against independently sourced clean transcriptions and routes every line through human review, combining automatic verification with expert oversight"

Permutation-equivariant architectures are motivated by the fact that graphs are invariant under node permutations

"Graphs are invariant under node permutations, motivating the use of permutation-equivariant architectures in generative models."

The natural Wasserstein geometry for graph pairs compared up to node relabeling is that of the graph quotient space

"once graph pairs are compared up to node relabeling, the natural Wasserstein geometry is that of the graph quotient space"

The Euclidean quotient metric coincides with the Gromov-Monge distance obtained by optimally relabeling nodes

"The Euclidean quotient metric of this space coincides with the Gromov--Monge distance, obtained by optimally relabeling the nodes."

The quantum-inspired representation can supply context-dependent parameters to classical car-following models and give autonomous vehicles a live behavioral read of surrounding drivers with short-horizon motion forecasts.

"it supplies context-dependent parameters to classical car-following models, and gives an autonomous vehicle a live behavioral read of the surrounding drivers with a short-horizon forecast of their motion."

Gemini 3.5 Transcribe provides more intelligent speech-to-text transcription capabilities

"you can get more intelligent speech-to-text transcription with Gemini 3.5 Transcribe"

Large multilingual foundation models like Whisper can be successfully adapted to extremely low-resource indigenous languages with very limited training data

"multilingual foundation models can be successfully adapted to extremely low-resource indigenous languages"

ASR technologies have achieved remarkable performance in recent years through large multilingual foundation models

"Automatic Speech Recognition (ASR) technologies have achieved remarkable performance in recent years through the use of large multilingual foundation models"

Fine-tuned Whisper Small model achieved 37.5% WER and 7.45% CER on Baniwa language with only 0.54 hours of training speech

"The best model achieved a WER of 37.5% and a CER of 7.45%, demonstrating that multilingual foundation models can be successfully adapted to extremely low-resource indigenous languages"

Deep neural networks often exploit spurious associations in their training data through shortcut learning

"Deep neural networks often exploit spurious associations in their training data, a failure known as shortcut learning."

ICON decomposition recovers concept importance more accurately than seven alternative baseline methods on synthetic data with known ground truth

"On synthetic data with known ground truth, ICON recovers concept importance more accurately than seven alternative baseline methods."

ICON decomposition can isolate concepts that models genuinely rely on and quantify unexplained representation in real-world imaging models

"On skin-lesion and brain-imaging models, it isolates the concepts on which a model genuinely relies, quantifies the representation unexplained by any of the supplied concepts, and yields sparse explanations that we validate by retraining and out-of-distribution testing."

Muon is a strong optimizer for matrix-valued parameters in large language model pretraining that approximately orthogonalizes momentum using Newton-Schulz iterations

"Muon has emerged as a strong optimizer for the matrix-valued parameters in large language model pretraining, approximately orthogonalizing its momentum with a few Newton-Schulz iterations."