Claims Browser

Fragment-based machine translation significantly outperforms standard k-shot and basic drafting methods across 6 languages and up to 5 domains per language

"Our experiments with the Qwen3 model family, over 6 languages, including up to 5 domains per language, demonstrate that fragment-based MT significantly outperforms alternative methods like standard k-shot or basic drafting."

Batch prompting makes LLM inference more efficient by processing multiple instances simultaneously but suffers from unpredictable downstream task performance

"batch prompting makes large language model inference more efficient by processing multiple instances simultaneously, it suffers from unpredictable downstream task performance"

Cascaded batch prompting resolves the unpredictability of conventional batch prompting by disentangling complex reasoning from symbol grounding

"cascaded batch prompting, a two-stage approach designed to resolve the unpredictability of conventional batch prompting by disentangling complex reasoning from symbol grounding"

Cascaded batch prompting outperforms standard single prompting baseline while achieving speedup proportional to batch size, establishing new state of the art on the Pareto frontier

"the proposed method outperforms the standard single prompting baseline while achieving a speedup proportional to batch size, establishing a new state of the art on the Pareto frontier"

Object-Conditioned Social Diffusion (OCSD) achieves state-of-the-art results on pedestrian trajectory forecasting benchmarks, reducing two-second path error by 31.3% on Humans in Kitchens and 33.2% on HOI-M3 compared to prior work

"OCSD achieves state-of-the-art results on the Humans in Kitchens (HiK) and HOI-M3 benchmarks. It reduces the two-second path error by 121.5 mm (31.3%) on HiK and 130.5 mm (33.2%) on HOI-M3 compared to prior work"

Integrating motion history, multi-person interactions, and object cues into a unified conditional diffusion framework enables more accurate forecasting of human movement in complex scenes

"OCSD uses an object-conditioning mechanism that modulates denoising at every timestep, enabling fine-grained human-object reasoning, and a social encoder that models the interactions between all humans in the scene"

OCSD produces more realistic long-term trajectory forecasts by naturally handling varying group sizes and complex social interactions

"our model naturally handles varying group sizes, complex social interactions, and supports sampling multiple plausible futures"

Multimodal representation learning is shifting from traditional two-tower architectures to LLM-based embedders due to their strong instruction-following capabilities

"Multimodal representation learning has been shifting from traditional two-tower architectures to large language model (LLM)-based embedders due to their strong instruction-following capabilities."

OmniUE is the first Omni-Interactive Universal Embedder that learns a unified embedding space across text, video, and audio and supports omni-interactive querying

"we propose the first Omni-Interactive Universal Embedder (OmniUE), which not only learns a unified embedding space across text, video, and audio by leveraging intermediate-layer representations from dedicated learnable tokens, but also supports omni-interactive querying, enabling users to provide inputs in the form of text, visual regions of interest, and audio spans."

OmniUE achieves 10.5% improvement on textual-interactive video benchmarks over state-of-the-art baselines

"OmniUE consistently surpasses state-of-the-art baselines across diverse modalities, with average improvements of 10.5% on textual-interactive video benchmarks (MMEB-v2-video)"

OmniUE achieves 83.7% improvement on visual-interactive benchmarks over state-of-the-art baselines

"OmniUE consistently surpasses state-of-the-art baselines across diverse modalities, with average improvements of 10.5% on textual-interactive video benchmarks (MMEB-v2-video), 1.1% on audio tasks (MAEB), 83.7% on visual-interactive benchmarks (SCaR), and 24.1% on our omni-interactive OmniCHOIR benchmark."

RL-style post-training can substantially improve chain-of-thought reasoning, long-horizon planning, and self-correction in language models

"such RL-style post-training ("RL-for-LLMs") can substantially improve chain-of-thought reasoning, long-horizon planning, and self-correction"

State-of-the-art reasoning language model training requires millions of GPU-hours

"state-of-the-art RLM training requires millions of GPU-hours"

Modern RLM training pipelines stress hardware far beyond classical supervised LLM training

"tightly coupled multi-model pipelines that stress modern hardware far beyond classical supervised LLM training"

Reliable assessment of causal research designs in social sciences has relied entirely on manual expert analysis until now

"Reliable assessment of causal research designs in the social sciences is critical for evidence-based policy-making, yet has so far relied entirely on manual expert analysis."

Passage length is the main driver of performance in automated research design assessment, explaining 52-66% of variance

"we find that passage length is the main driver of performance, explaining 52-66% of the variance."

Human and machine difficulty do not align in research design assessment - designs hardest for AI systems are not those where expert annotators disagree most

"the designs that prove hardest for the system are not those on which expert annotators disagree most, pointing to two independent sources of task difficulty."

Diffusion models can be trained to learn the mapping between parameter space and observable space for solving inverse problems

"A DM is trained to learn the mapping between the parameter space and the observable space."

The EU AI Act's high-risk obligations have applied since August 2, 2026

"the EU AI Act, whose high-risk obligations have applied since 2 August 2026"