Discusses handling intercurrent events in clinical trials with time-to-event outcomes.
problem Handling intercurrent events in clinical trials with time-to-event outcomes.
method Defines estimands and six ICE handling strategies, including new competing-risk strategy.
result Novel methods for handling intercurrent events in clinical trials with time-to-event outcomes.
Trading-R1 uses LLMs for financial trading, improving risk-adjusted returns.
problem Lack of interpretability and trust in AI for finance.
method Supervised fine-tuning and reinforcement learning with a curriculum.
result Improved risk-adjusted returns and lower drawdowns compared to other models.
Alpha-R1 uses LLMs to reason about economic factors and news for better alpha screening.
problem Challenges in data-driven investment strategies due to signal decay and regime shifts.
method Reinforcement learning trained on 8B parameters to evaluate alpha relevance under changing market conditions.
result Empirically outperforms benchmark strategies and shows improved robustness to alpha decay.
Trade-R1 bridges verifiable rewards to stochastic financial markets via process-level reasoning verification.
problem Extending RL to financial markets where rewards are verifiable but noisy.
method A verification method that transforms reasoning over financial documents into a structured RAG task, using a triangular consistency metric.
result DSR achieves superior cross-market generalization while maintaining reasoning consistency.
Unified framework for understanding GRPO as U-statistic.
problem Theoretical properties of GRPO remain less studied.
method Unified framework through classical U-statistics.
result GRPO is asymptotically equivalent to an oracle policy gradient algorithm.
Proposes a strategy to train models with minimal labeled data.
problem Scarce and expensive labeled data for medical tasks.
method Recursive training strategy to use image-level annotations for pixel-level segmentation.
result Improved segmentation of intracranial hemorrhage in CT scans.
This article has two purposes. In \cite{R3} (math.KT/0405211) we showed that the FIC (Fibered Isomorphism Conjecture for pseudoisotopy functor) for a particular class of 3-manifolds (we denoted this class by \cal C) is the key to prove the FIC for 3-manifold groups in general. And we proved the FIC for the fundamental …
Validation is one of the most important aspects of clustering, but most approaches have been batch methods. Recently, interest has grown in providing incremental alternatives. This paper extends the incremental cluster validity index (iCVI) family to include incremental versions of Calinski-Harabasz (iCH), I index and …
This note has an experimental nature and contains no new theorems. We introduce certain moves for classical knot diagrams that for all the very many examples we have tested them on give a monotonic complete simplification. A complete simplification of a knot diagram D is a sequence of moves that transform D into a diag…
SNAPO optimizes policies for complex sequential decisions using differentiable simulation.
problem Optimizing policies for high-dimensional, sequential decisions under uncertainty.
method Embeds neural policy in a differentiable simulator, computes gradients efficiently.
result Produces sensitivities at a cost proportional to one reverse pass, regardless of sensitivity count.
Study evaluates consistency of LLMs in binary text classification, providing systematic guidance.
problem Lack of reliable methods for evaluating large language model (LLM) binary text classification.
method Adapting psychometric principles, the study determines sample size requirements, develops metrics for invalid responses, and evaluates intra- and inter-rater reliability.
result LLMs demonstrated high intra-rater consistency, achieving perfect agreement on 90-98% of examples, with smaller models outperforming larger counterparts.
A minimalist approach improves LLM reasoning by filtering incorrect responses.
problem Improving large language model (LLM) reasoning on complex tasks.
method Revisit GRPO from a reinforce-like algorithm perspective, proposing Reinforce-Rej.
result RAFT, a simple rejection sampling baseline, outperforms GRPO and PPO.
Enhances large language models' reasoning through simpler off-policy reinforcement learning.
problem Improving large language models' ability to reason and solve problems.
method EM Policy Gradient, optimizing expected return over reasoning trajectories using Expectation-Maximization (EM) optimization.
result Achieves comparable or slightly superior performance to state-of-the-art methods on reasoning datasets, with additional cognitive behaviors.
The study examines how extra compute during testing affects the performance of large language models.
problem Understanding the conditions under which test-time scaling improves model performance.
method An in-context weight prediction task for linear regression was used to train transformers. The performance was analyzed under varying levels of test-time compute.
result Training transformers on diverse, relevant, and hard tasks leads to the best performance for test-time scaling.
Study proposes a multi-agent system using LLMs for REIT trading, outperforming benchmarks.
problem Low-volatility Chinese REIT market, low risk-adjusted returns.
method Multi-agent framework with four types of agents, prediction model pathways, fine-tuning.
result Multi-agent strategies outperform buy-and-hold in terms of return, Sharpe ratio, and drawdown.
Painless Activation Steering automates post-training for LMs without manual intervention.
problem Manual post-training methods are time-consuming and labor-intensive.
method Painless Activation Steering (PAS) is a fully automated approach that requires no manual intervention.
result PAS reliably improves performance for behavior tasks but not for intelligence-oriented tasks.
This paper proposes exact and approximation algorithms for Sparse PCA, improving interpretability and scalability.
problem Selecting a prespecified-size principal submatrix from a covariance matrix to maximize its largest eigenvalue.
method Proposes two exact mixed-integer SDPs and a mixed-integer linear program (MILP) for SPCA, analyzes theoretical optimality gaps, and develops approximation algorithms.
result The proposed algorithms achieve strong theoretical optimality and effective scalability, with continuous relaxations close to optimality and MILP solving small to medium-size instances.
In this paper, we introduce a new learning strategy based on a seminal idea of Mojirsheibani (1999, 2000, 2002a, 2002b), who proposed a smart method for combining several classifiers, relying on a consensus notion. In many aggregation methods, the prediction for a new observation x is computed by building a linear or c…