Self-play post-training methods has emerged as an effective approach for finetuning large language models and turn the weak language model into strong language model without preference data. However, the theoretical foundations for self-play finetuning remain underexplored. In this work, we tackle this by connecting self-play finetuning with adversarial imitation learning by formulating finetuning procedure as a min-max game between the model and a regularized implicit reward player parameterized by the model itself. This perspective unifies self-play imitation and […]

Ver mais

arXiv:2604.18775v1 Announce Type: new Abstract: Detecting jailbreak behaviour in large language models remains challenging, particularly when strongly aligned models produce harmful outputs only rarely. In this work, we present an empirical study of output based jailbreak detection under realistic conditions using the JailbreakBench Behaviors dataset and multiple generator models with varying alignment strengths. We evaluate both a lexical TF-IDF detector and a generation inconsistency based detector across different sampling budgets. Our results show that single output evaluation systematically […]

Ver mais

Obtaining labelled ECG data for developing supervised models is challenging. Contrastive learning (CL) has emerged as a promising pretraining approach that enables effective transfer learning with limited labelled data. However, existing CL frameworks either focus solely on global context or fail to exploit ECG-specific characteristics. Furthermore, these methods rely on hard contrastive targets, which may not adequately capture the continuous nature of feature similarity in ECG signals. In this paper, we propose Beat-SSL, a contrastive learning framework that […]

Ver mais

arXiv:2603.12414v1 Announce Type: new Abstract: State Space Models (SSMs) such as Mamba achieve linear-time sequence processing through input-dependent recurrence, but this mechanism introduces a critical safety vulnerability. We show that the spectral radius rho(A-bar) of the discretized transition operator governs effective memory horizon: when an adversary drives rho toward zero through gradient-based Hidden State Poisoning, memory collapses from millions of tokens to mere dozens, silently destroying reasoning capacity without triggering output-level alarms. We prove an Evasion Existence Theorem […]

Ver mais

I think we will see some people employed (though perhaps not explicitly) as meat shields: people who are accountable for ML systems under their supervision. The accountability may be purely internal, as when Meta hires human beings to review the decisions of automated moderation systems. It may be external, as when lawyers are penalized for submitting LLM lies to the court. It may involve formalized responsibility, like a Data Protection Officer. It may be convenient for a company […]

Ver mais

Lightweight LLM for converting text to structured data Novel training procedure and decoding mechanism enable model to outperform much larger foundation model prompted to perform the same task. Conversational AI Karim Bouyarmane February 06, 10:14 AM February 06, 10:27 AM One of the most important features of todays generative models is their ability to take unstructured, partially unstructured, or poorly structured inputs and convert them into structured objects that conform to specific schemas relational-database fixed schemas, document store […]

Ver mais

arXiv:2603.20199v1 Announce Type: new Abstract: Sticky notes remain a durable collaborative medium because they support rapid idea externalization, rearrangement, and coordination of group attention through spatial organization while being low-friction and lightweight. Recent AR systems suggest new ways to externalize ideas in shared physical space, including spatial annotations and digital workspaces. We introduce AnchorNote, a co-located AR system that lets collaborators intentionally capture spoken ideas as spatially anchored sticky notes via live transcription and LLM summarization. We evaluated […]

Ver mais

How are you, hacker? 🪐Want to know what’s trending right now?: The Techbeat by HackerNoon has got you covered with fresh content from our trending stories of the day! Set email preference here. ## SERP Benchmarks: Success Rates and Latency at Scale By @brightdata [ 8 Min read ] ​​We benchmark SERP APIs for success rate, ​​speed, and stability under load. Learn which setup delivers consistent results for AI agents ​​and deep research. Read More. MEXC Reports 2.35 […]

Ver mais

arXiv:2601.08838v1 Announce Type: new Abstract: Large-scale Text-to-SQL benchmarks such as BIRD typically assume complete and accurate database annotations as well as readily available external knowledge, which fails to reflect common industrial settings where annotations are missing, incomplete, or erroneous. This mismatch substantially limits the real-world applicability of state-of-the-art (SOTA) Text-to-SQL systems. To bridge this gap, we explore a database-centric approach that leverages intrinsic, fine-grained information residing in relational databases to construct missing evidence and improve Text-to-SQL accuracy under […]

Ver mais