Combinatorial optimization (CO) problems, central to operation research and theoretical computer science, present significant computational challenges due to their NP-hard nature. While large language models (LLMs) have emerged as promising tools for CO--either by directly generating solutions or synthesizing solver-specific codes--existing approaches often neglect critical structural priors inherent to CO problems, leading to suboptimality and iterative inefficiency. Inspired by human experts' success in leveraging CO structures for algorithm design, we propose STRCMP, a novel structure-aware LLM-based algorithm discovery framework that systematically integrates structure priors to enhance solution quality and solving efficiency. Our framework combines a graph neural network (GNN) for extracting structural embeddings from CO instances with an LLM conditioned on these embeddings to identify high-performing algorithms in the form of solver-specific codes. This composite architecture ensures syntactic correctness, preserves problem topology, and aligns with natural language objectives, while an evolutionary refinement process iteratively optimizes generated algorithm. Extensive evaluations across Mixed Integer Linear Programming and Boolean Satisfiability problems, using nine benchmark datasets, demonstrate that our proposed STRCMP outperforms five strong neural and LLM-based methods by a large margin, in terms of both solution optimality and computational efficiency. The code and learned model will be publicly available upon the acceptance of the paper.

Recent advances in Foundation Models for Materials Science are poised to revolutionize the discovery, manufacture, and design of novel materials with tailored properties and responses. Although great strides have been made, successes have been restricted to materials classes where multi-million sample data repositories can be readily curated (e.g., atomistic structures). Unfortunately, for many structural and functional materials (e.g., mesoscale structured metal alloys), such datasets are too costly or prohibitive to construct; instead, datasets are limited to very few examples. To address this challenge, we introduce a novel machine learning approach for learning from hyper-sparse, complex spatial data in scientific domains. Our core contribution is a physics-driven data augmentation scheme that leverages an ensemble of local generative models, trained on as few as five experimental observations, and coordinates them through a novel diversity curation strategy to generate a large-scale, physically diverse dataset. We utilize this framework to construct PolyMicros, the first Foundation Model for polycrystalline materials (a structural material class important across a broad range of industrial and scientific applications). We demonstrate the utility of PolyMicros by zero-shot solving several long standing challenges related to accelerating 3D experimental microscopy. Finally, we make both our models and datasets openly available to the community.

Electricity price forecasting is a critical component of modern energy-management systems, yet existing approaches heavily rely on numerical histories and ignore contemporaneous textual signals. We introduce NSW-EPNews, the first benchmark that jointly evaluates time-series models and large language models (LLMs) on real-world electricity-price prediction. The dataset includes over 175,000 half-hourly spot prices from New South Wales, Australia (2015-2024), daily temperature readings, and curated market-news summaries from WattClarity. We frame the task as 48-step-ahead forecasting, using multimodal input, including lagged prices, vectorized news and weather features for classical models, and prompt-engineered structured contexts for LLMs. Our datasets yields 3.6k multimodal prompt-output pairs for LLM evaluation using specific templates. Through compresive benchmark design, we identify that for traditional statistical and machine learning models, the benefits gain is marginal from news feature. For state-of-the-art LLMs, such as GPT-4o and Gemini 1.5 Pro, we observe modest performance increase while it also produce frequent hallucinations such as fabricated and malformed price sequences. NSW-EPNews provides a rigorous testbed for evaluating grounded numerical reasoning in multimodal settings, and highlights a critical gap between current LLM capabilities and the demands of high-stakes energy forecasting.

Confidence estimation is crucial for reflecting the reliability of large language models (LLMs), particularly in the widely used closed-source models. Utilizing data augmentation for confidence estimation is viable, but discussions focus on specific augmentation techniques, limiting its potential. We study the impact of different data augmentation methods on confidence estimation. Our findings indicate that data augmentation strategies can achieve better performance and mitigate the impact of overconfidence. We investigate the influential factors related to this and discover that, while preserving semantic information, greater data diversity enhances the effectiveness of augmentation. Furthermore, the impact of different augmentation strategies varies across different range of application. Considering parameter transferability and usability, the random combination of augmentations is a promising choice.

Pretrained Foundation Models (PFMs) have transformed numerous applications by enabling efficient adaptation to customized tasks. Parameter-Efficient Fine-Tuning (PEFT) has emerged as a resource-efficient alternative to full fine-tuning, especially leveraging reparameterized weights $ΔW$ to adapt models for downstream tasks. However, a critical yet underexplored question remains: can we utilize well-pretrained weights $W_0$ to guide the update of task-specific $ΔW$, avoiding inefficient training it from scratch? To end this, we propose Generative Parameter-Efficient Fine-Tuning (GenFT), a novel method that extracts structured, transferable information from $W_0$ for efficient $ΔW$ training. To extract row and column structure information, GenFT applies row and column transformations to distill essential patterns from $W_0$. A tailored policy further decomposes $ΔW$ into layer-shared and layer-specific components, balancing information reuse and individualized flexibility. GenFT is simple yet effective, achieving superior performance across CV and NLP tasks. Extensive experiments on VTAB-1K, FGVC, and GLUE benchmarks demonstrate that GenFT outperforms state-of-the-art PEFT methods, offering a new perspective for efficient model adaptation.

Thousands of university students in the UK have been caught misusing ChatGPT and other artificial intelligence tools in recent years, while traditional forms of plagiarism show a marked decline, a Guardian investigation can reveal. A survey of academic integrity violations found almost 7,000 proven cases of cheating using AI tools in 2023-24, equivalent to 5.1 for every 1,000 students. That was up from 1.6 cases per 1,000 in 2022-23. Figures up to May suggest that number will increase again this year to about 7.5 proven cases per 1,000 students – but recorded cases represent only the tip of the iceberg, according to experts. The data highlights a rapidly evolving challenge for universities: trying to adapt assessment methods to the advent of technologies such as ChatGPT and other AI-powered writing tools.

Are machines about to become smarter than humans? If you take the leaders of artificial intelligence companies at their word, their products mean that the coming decade will be quite unlike any in human history: a golden era of "radical abundance", where high-energy physics is "solved" and we see the beginning of space colonisation. But researchers working with today's most powerful AI systems are finding a different reality, in which even the best models are failing to solve basic puzzles that most humans find trivial, while the promise of AI that can "reason" seems to be overblown. So, whom should you believe? Sam Altman and Demis Hassabis, the CEOs of OpenAI and Google DeepMind, respectively, have both made recent claims that powerful, world-altering AI systems are just around the corner.

Welcome to The Full Nerd newsletter--your weekly dose of hardcore hardware talk from the enthusiasts at PCWorld. In it, we dive into the hottest topics from our YouTube show, plus interesting news from across the web. Attending the Nintendo Switch 2 launch at our local Nintendo Store felled both Adam and Will, delaying our usual Tuesday episode. But don't worry: I still have plenty of juicy news bits to share with you below. Also our Micro Center tour videos are live!

For more than 20 years, print media has been a bit of a punching bag for digital-technology companies. Craigslist killed the paid classifieds, free websites led people to think newspapers and magazines were committing robbery when they charged for subscriptions, and the smartphone and social media turned reading full-length articles into a chore. Now generative AI is in the mix--and many publishers, desperate to avoid being left behind once more, are rushing to harness the technology themselves. Several major publications, including The Atlantic, have entered into corporate partnerships with OpenAI and other AI firms. Any number of experiments have ensued--publishers have used the software to help translate work into different languages, draft headlines, and write summaries or even articles.

The dominant way of judging Large Language Models (LLMs) has been to ask how well they can recall explicit facts from very long inputs. While today's best models achieve near perfect recall, this masks a harder skill: performing multi-step reasoning and tracking intermediate state that never appears verbatim. We introduce Verbose ListOps (VLO), a benchmark that embeds deterministic ListOps computations inside narrative camouflage and, crucially, allows step-level evaluation of every intermediate result. Experiments show that models which solve raw ListOps with approximately 100% accuracy collapse on VLO after only 10,000 tokens. By exposing where a model's reasoning chain first diverges, VLO moves assessment beyond sheer context length and toward genuine comprehension. VLO's generation pipeline is task-agnostic: it can weave any deterministically verifiable reasoning schema -- arithmetic, symbolic, abductive, inductive or defeasible -- into narrative form. This makes VLO a reusable test-bed for the next wave of reasoning-centric model designs, not merely those with step-explicit scaffolds.