The edifice of native Vision-Language Models (VLMs) has emerged as a rising contender to typical modular VLMs, shaped by evolving model architectures and training paradigms. Yet, two lingering clouds cast shadows over its widespread exploration and promotion: (-) What fundamental constraints set native VLMs apart from modular ones, and to what extent can these barriers be overcome? (-) How to make research in native VLMs more accessible and democratized, thereby accelerating progress in the field. In this paper, we clarify these challenges and outline guiding principles for constructing native VLMs. Specifically, one native VLM primitive should: (i) effectively align pixel and word representations within a shared semantic space; (ii) seamlessly integrate the strengths of formerly separate vision and language modules; (iii) inherently embody various cross-modal properties that support unified vision-language encoding, aligning, and reasoning. Hence, we launch NEO, a novel family of native VLMs built from first principles, capable of rivaling top-tier modular counterparts across diverse real-world scenarios. With only 390M image-text examples, NEO efficiently develops visual perception from scratch while mitigating vision-language conflicts inside a dense and monolithic model crafted from our elaborate primitives. We position NEO as a cornerstone for scalable and powerful native VLMs, paired with a rich set of reusable components that foster a cost-effective and extensible ecosystem. Our code and models are publicly available at: https://github.com/EvolvingLMMs-Lab/NEO.

As Large Language Models (LLMs) become increasingly integrated into software development workflows, they also become prime targets for adversarial attacks. Among these, backdoor attacks are a significant threat, allowing attackers to manipulate model outputs through hidden triggers embedded in training data. Detecting such backdoors remains a challenge, and one promising approach is the use of Spectral Signature defense methods that identify poisoned data by analyzing feature representations through eigenvectors. While some prior works have explored Spectral Signatures for backdoor detection in neural networks, recent studies suggest that these methods may not be optimally effective for code models. In this paper, we revisit the applicability of Spectral Signature-based defenses in the context of backdoor attacks on code models. We systematically evaluate their effectiveness under various attack scenarios and defense configurations, analyzing their strengths and limitations. We found that the widely used setting of Spectral Signature in code backdoor detection is often suboptimal. Hence, we explored the impact of different settings of the key factors. We discovered a new proxy metric that can more accurately estimate the actual performance of Spectral Signature without model retraining after the defense.

We present Entropic Mutual-Information Geometry Large-Language Model Alignment (ENIGMA), a novel approach to Large-Language Model (LLM) training that jointly improves reasoning, alignment and robustness by treating an organisation's policies/principles as directions to move on a model's information manifold. Our single-loop trainer combines Group-Relative Policy Optimisation (GRPO), an on-policy, critic-free RL method with Chain-of-Thought (CoT)-format only rewards; a Self-Supervised Alignment with Mutual Information (SAMI)-style symmetric InfoNCE auxiliary; and an entropic Sinkhorn optimal-transport regulariser on hidden-state distributions to bound geometry drift. We also introduce infoNCE metrics that specialise to a standard MI lower bound under matched negatives to measure how strongly a model's CoT encodes these policies. These metrics include a Sufficiency Index (SI) that enables the selection and creation of principles that maximise downstream performance prior to training. In our experiments using small (1B) LLMs, high-SI principles predict steadier training dynamics and improved benchmark performance over GRPO ablations. Our information-geometry analysis of trained models validates desirable structural change in the manifold. These results support our hypothesis that reasoning, alignment, and robustness are projections of a single information-geometric objective, and that models trained using ENIGMA demonstrate principled reasoning without the use of a reward model, offering a path to trusted capability

Large Language Models (LLMs) are increasingly integral to information dissemination and decision-making processes. Given their growing societal influence, understanding potential biases, particularly within the political domain, is crucial to prevent undue influence on public opinion and democratic processes. This work investigates political bias and stereotype propagation across eight prominent LLMs using the two-dimensional Political Compass Test (PCT). Initially, the PCT is employed to assess the inherent political leanings of these models. Subsequently, persona prompting with the PCT is used to explore explicit stereotypes across various social dimensions. In a final step, implicit stereotypes are uncovered by evaluating models with multilingual versions of the PCT. Key findings reveal a consistent left-leaning political alignment across all investigated models. Furthermore, while the nature and extent of stereotypes vary considerably between models, implicit stereotypes elicited through language variation are more pronounced than those identified via explicit persona prompting. Interestingly, for most models, implicit and explicit stereotypes show a notable alignment, suggesting a degree of transparency or "awareness" regarding their inherent biases. This study underscores the complex interplay of political bias and stereotypes in LLMs.

The typical offline protocol to evaluate recommendation algorithms is to collect a dataset of user-item interactions and then use a part of this dataset to train a model, and the remaining data to measure how closely the model recommendations match the observed user interactions. This protocol is straightforward, useful and practical, but it only captures performance of a particular model trained at some point in the past. We know, however, that online systems evolve over time. In general, it is a good idea that models reflect such changes, so models are frequently retrained with recent data. But if this is the case, to what extent can we trust previous evaluations? How will a model perform when a different pattern (re)emerges? In this paper we propose a methodology to study how recommendation models behave when they are retrained. The idea is to profile algorithms according to their ability to, on the one hand, retain past patterns - stability - and, on the other hand, (quickly) adapt to changes - plasticity. We devise an offline evaluation protocol that provides detail on the long-term behavior of models, and that is agnostic to datasets, algorithms and metrics. To illustrate the potential of this framework, we present preliminary results of three different types of algorithms on the GoodReads dataset that suggest different stability and plasticity profiles depending on the algorithmic technique, and a possible trade-off between stability and plasticity. Although additional experiments will be necessary to confirm these observations, they already illustrate the usefulness of the proposed framework to gain insights on the long term dynamics of recommendation models.

Sam Fender has won the 2025 Mercury Prize for his third album, People Watching, a steely-eyed dissection of working-class life in the north of England. The singer looked stunned when his name was announced. I didn't think that was going to happen at all, he told the BBC as he came off stage. I've spent the last 10 minutes crying. Fender beat the likes of Pulp and Wolf Alice - both former winners of the £25,000 prize for the best British or Irish album of the year - at a star-studded ceremony in Newcastle's Utilita Arena.

Trump says he will meet Putin in Hungary for Ukraine talks after'very productive' call US President Donald Trump says great progress was made during a phone call with Russian President Vladimir Putin on Thursday, with the pair agreeing to face-to-face talks in Hungary. He said the call, the first with Putin since mid-August, was very productive, adding that teams from Washington and Moscow will meet next week. Trump did not confirm a date for his meeting with Putin in Budapest. The Kremlin said work on the summit would begin immediately after the extremely frank and trustful call. The talks came a day before Ukraine's President Zelensky was to visit the White House, and with Trump weighing whether to arm Ukraine with Tomahawk missiles capable of striking deep into Russia.

EU foreign policy chief Kaja Kallas has said a new anti-drone system should be fully operational by the end of 2027, as part of a drive to toughen defences against Russia and be fully prepared for possible conflict by 2030. Drones are already redefining warfare. Having drone defences is no longer optional for anyone, Kallas said, referring to Russia's ongoing war in Ukraine and fears that Moscow may attack the EU. The European Commission's defence roadmap also proposes strengthening the EU's eastern borders and building air and space shields. Several EU nations have faced Russian incursions into their airspace and US President Donald Trump has urged the bloc to do more to defend itself.

Plans for a groundbreaking rare earths refinery in East Yorkshire have been abandoned, after the company behind the project decided to seek investment in the United States instead. Pensana has spent the past seven years developing a rare earths mine in Angola. The $268m (£185m) project, one of the largest of its kind in the world, will begin delivering raw materials in 2027. The company had planned to build a refinery at the Saltend Chemicals Plant near Hull, which would have processed the raw materials into metals used to create powerful magnets. These magnets would then be used in high-tech applications such as motors for electric vehicles, wind turbines and robotics.

Sharon Osbourne has said it would be amazing if Birmingham Airport was renamed in honour of her late husband, rock legend Ozzy Osbourne. The TV personality has given her support to a campaign to call the airport Ozzy Osbourne International, which was launched by podcaster and comedian Dan Hudson after the Black Sabbath singer died at the age of 76 in July. More than 70,000 people have signed a petition backing the idea, which Hudson said was inspired by airports being named after famous figures such as John Lennon. It would be amazing, Osbourne said of a potential rebrand. It's just a dream right now, but sometimes dreams come true.