26838 research outputs found
Sort by
Position: unlocking the full potential of data science requires tabular foundation models, agents, and humans
Get PDFDespite its vast potential, data science remains constrained by manual workflows and fragmented tools. Meanwhile, foundation models have transformed natural language and computer vision - and are beginning to bring similar breakthroughs to structured data, particularly the ubiquitous tabular data central to data science. At the same time, there are strong claims that fully autonomous agentic data science systems will emerge. We argue that, rather than replacing data scientists, the future of data science lies in a new paradigm that amplifies their impact: collaborative systems that tightly integrate agents and tabular foundation models (TFMs) with human experts. In this paper, we discuss the potential and challenges of navigating the interplay between these three and present a research agenda to guide this disruption toward a more accessible, robust, and human-centered data science
QMA = QMA1 with an infinite counter
Get PDFA long-standing open problem in quantum complexity theory is whether , the quantum analogue of , is equal to , its one-sided error variant. We show that , where is like , but the verifier has an infinite register, as part of their witness system, in which they can efficiently perform a shift (increment) operation. We call this register an "infinite counter", and compare it to a program counter in a Las Vegas algorithm. The result means such an infinite register does not increase the power of , but does imply perfect completeness.
By truncating our construction to finite dimensions, we get a -amplifier that only amplifies completeness, not soundness, but does so in significantly less time than previous amplifiers. Our new construction achieves completeness using calls to each of the original verifier and its inverse, and other gates, proving that has completeness exponentially close to 1, i.e. for any polynomial
SQaLe: A text-to-SQL dataset generation pipeline grounded in real schemas
No full textComposable pipeline for curating large-scale text-to-SQL corpora by extending database schemas, synthesising natural-language questions, and validating SQL programs with LLMs.
The dataset can be accessed under trl-lab/SQaLe-text-to-SQL-dataset/ on Hugging Face Datasets
Combining style and semantics for robust authorship verification
Get PDFAuthorship Verification is a key task in Natural Language Processing, essential for applications like plagiarism detection and content authentication. This paper analyzes the use of deep learning models for Authorship Verification, focusing on combining semantic and style features to enhance model performance. We propose three models: the Feature Interaction Network, Pairwise Concatenation Network, and Siamese Network, which aim to determine if two texts are written by the same author. Each model uses RoBERTa embeddings to capture semantic content and incorporates style features such as sentence length, word frequency, and punctuation to differentiate authors based on writing style.
Our results confirm that incorporating style features consistently improves model performance, with the extent of improvement varying by architecture. This demonstrates the value of combining semantic and stylistic information for Authorship Verification. While limitations such as RoBERTa’s fixed input length and the use of predefined style features exist, they do not hinder model effectiveness and point to clear opportunities for future enhancement through extended input handling and dynamic style feature extraction.
In contrast to prior studies such as Bevendorff et al., (2020) and Kestemont, et al., (2022), which relied on balanced and homogeneous datasets with consistent topics and well-formed language, our work evaluates models on a more challenging, imbalanced, and stylistically diverse dataset, better reflecting real-world Authorship Verification conditions. Despite the increased difficulty, our models achieve competitive results, underscoring their robustness and practical applicability.
These findings support the value of combining semantic and style features for real-world Authorship Verification
Rao-Blackwellized e-variables
Get PDFWe show that for any concave utility, the expected utility of an e-variable can only increase after conditioning on a sufficient statistic. The simplest form of the result has an extremely straightforward proof, which follows from a single application of Jensen's inequality. Similar statements hold for compound e-variables, asymptotic e-variables, and e-processes. These results echo the Rao-Blackwell theorem, which states that the expected squared error of an estimator can only decrease after conditioning on a sufficient statistic. We provide several applications of this insight, including a simplified derivation of the log-optimal e-variable for linear regression with known variance
Security analysis of covercrypt: A quantum-safe hybrid key encapsulation mechanism for hidden access policies
Get PDFThe ETSI Technical Specification 104 015 proposes a framework to build Key Encapsulation Mechanisms (KEMs) with access policies and attributes, in the Ciphertext-Policy Attribute-Based Encryption (CP-ABE) vein. Several security guarantees and functionalities are claimed, such as pre-quantum and post-quantum hybridization to achieve security against Chosen-Ciphertext Attacks (CCA), anonymity, and traceability. In this paper, we present a formal security analysis of a more generic construction, with application to the specific Covercrypt scheme, based on the pre-quantum ECDH and the post-quantum ML-KEM Key Encapsulation Mechanisms. We additionally provide an open-source library that implements the ETSI standard, in Rust, with high efficiency
[COMP24] The Automated Negotiating Agents Competition (ANAC) 2024 Challenges and Results
Get PDFThis paper introduces the main research challenges and results of the 15th International Automated Negotiating Agents Competition (ANAC 2024). The main challenges addressed are learning the reservation value in bilateral negotiation and designing a factory agent employing concurrent negotiation in supply chain management. Additionally, it outlines the future directions for the competition