bilinear_family_forced

"For continuous nonconstant functions F:R>0→R with F(1)=0 satisfying the composition law with a symmetric polynomial P of degree at most two, the combiner is necessarily of the form P(u,v)=2u+2v+c uv, c∈R (Theorem 3.3.)."

"an inclusion of nonadditivity within an optimal effective Yukawa model gives better predictions... whose only free parameter is the AB cross-term"

"This report describes the physics case, the resulting detector requirements, and the evolving detector concepts for the experimental program at the Electron-Ion Collider (EIC)."

"Sparrow is preferred more often than baselines while being more resilient to adversarial probing by humans, violating our rules only 8% of the time when probed."

"We demonstrate that TD-MPC2 improves significantly over baselines across 104 online RL tasks spanning 4 diverse task domains, achieving consistently strong results with a single set of hyperparameters."

"log(pX(x)) = log(pH(f(x))) + log(|det ∂f(x)/∂x|)"

"One major advantage of the algorithm is that it requires hand-tuning of only 1 or 2 parameters compared to ∼N² for a traditional algorithm in an N-dimensional parameter space."

"In an analogy to complexity theory, debate with optimal play can answer any question in PSPACE given polynomial time judges (direct judging answers only NP questions)."

"cutout can be used in conjunction with existing forms of data augmentation and other regularizers"

"In addition, we present a historical overview of the evolution of LLMs for code generation and offer an empirical comparison using the HumanEval, MBPP, and BigCodeBench benchmarks"

"A novel lookup-free quantization approach that enables improving the visual generation quality of language models by learning a large vocabulary."

"We show how classifier-guided sampling and image inpainting can be reinterpreted as coherent planning strategies, and demonstrate the effectiveness of our framework in control settings that emphasize long-horizon decision-making and test-time flexibility."

"We present measurements of the temperature and E-mode polarization angular power spectra of the cosmic microwave background (CMB) from observations of 4% of the sky with SPT-3G... Combining our TT/TE/EE spectra with previously published SPT-3G CMB lensing results, we find parameters for the standard LCDM model consistent with Planck and ACT-DR6"

"synthetic cold-start data generation to produce code-augmented long-form reasoning traces"

"we propose R1-Searcher, a novel two-stage outcome-based RL approach designed to enhance the search capabilities of LLMs"

"we propose 3D-VLA by introducing a new family of embodied foundation models that seamlessly link 3D perception, reasoning, and action through a generative world model. Specifically, 3D-VLA is built on top of a 3D-based large language model (LLM), and a set of interaction tokens is introduced to engage with the embodied environment."

"Despite the simplicity in formulation and ease of training, our architecture is able to generate samples competitive with state-of-the-art GAN models for video generation on the BAIR Robot dataset"

"To assess the effectiveness of this cascading procedure... we construct a new image dataset, LSUN. It contains around one million labeled images for each of 10 scene categories and 20 object categories"

"a UWM integrates an action diffusion process and a video diffusion process within a unified transformer architecture, where independent diffusion timesteps govern each modality. By controlling each diffusion timestep, UWM can flexibly represent a policy, a forward dynamics, an inverse dynamics, and a video generator."

"We find that the ACT angular power spectra … are well fit by the sum of CMB and foregrounds, where the CMB spectra are described by the ΛCDM model. Combining ACT with larger-scale Planck data, the joint P-ACT dataset provides tight limits on the ingredients, expansion rate, and initial conditions of the universe … Ω_b h² = 0.0226 ± 0.0001, Ω_c h² = 0.118 ± 0.001, H_0 = 68.22 ± 0.36 km/s/Mpc, n_s = 0.974 ± 0.003, and σ_8 = 0.813 ± 0.005."

"We extend SAC to incorporate a number of modifications that accelerate training and improve stability with respect to the hyperparameters, including a constrained formulation that automatically tunes the temperature hyperparameter."

"Bilby provides expert-level parameter estimation infrastructure with straightforward syntax... for the analysis of compact binary mergers and other types of signal model"

"Any observation suggesting otherwise would be an indication of beyond-the-standard-model physics. Null results strengthen and quantify the Kerr black hole paradigm."

"We argue that language models hallucinate because the training and evaluation procedures reward guessing over acknowledging uncertainty"

"we present two parameter-reduction techniques to lower memory consumption and increase the training speed of BERT... our best model establishes new state-of-the-art results on the GLUE, RACE, and SQuAD benchmarks while having fewer parameters compared to BERT-large."

"Multi-task learning (MTL) has led to successes in many applications of machine learning, from natural language processing and speech recognition to computer vision and drug discovery. This article aims to give a general overview of MTL, particularly in deep neural networks. It introduces the two most common methods for MTL in Deep Learning, gives an overview of the literature, and discusses recent advances."

"VAPO, built on the Qwen 32B pre-trained model, attains a state-of-the-art score of 60.4. In direct comparison under identical experimental settings, VAPO outperforms the previously reported results of DeepSeek-R1-Zero-Qwen-32B and DAPO by more than 10 points."

"combining any three independent approaches to measure H0 in the late universe yields tension with the early Universe values between 4.0σ and 5.8σ"

"Agentic RAG transcends these limitations by embedding autonomous AI agents into the RAG pipeline. These agents leverage agentic design patterns reflection, planning, tool use, and multi-agent collaboration"

"Our approach incorporates new techniques for representation learning, optimization, and augmentation, enabling EVA-CLIP to achieve superior performance compared to previous CLIP models with the same number of parameters but significantly smaller training costs."