Models

Prompt Claude, ChatGPT, Gemini, or any other popular large language model (LLM) with a question like “What is the best film ever made?” and the response will vary, and you (and most worryingly, the people who built the LLM) have little idea exactly how it came up with that specific answer. This mysterious behavior can be useful in some situations. But—as a recent incident where OpenAI could not explain why its advanced pre-release model hacked AI company Hugging Face highlighted—it can have negative and alarming consequences too. And when frontier AI models are writing code, generating results humans could not achieve alone, and performing other important tasks across society, the need to interpret AI ‘thinking’ and outputs has never been greater. Goodfire, an AI lab focused solely on this very problem, recently made its cutting-edge Silico platform, filled with tools to interpret the behavior of AI, generally available to the public. As part of this, the company recently announced a new grant program offering $1 million in free Silico usage for academic and nonprofit interpretability researchers. These efforts aim to democratize AI interpretability, placing techniques previously available to a clutch of elite labs into the hands of ambitious research teams and startups that want to build and understand their own models or adapt open-source models for different purposes. Mechanistic interpretability Founded in 2024 and based in San Francisco, Goodfire aims to provide the tools that build the next generation of safe and powerful AI by understanding the structures inside them instead of treating AI models as black boxes. “Treating models like black boxes isn’t inevitable, it’s a choice,” says Eric Ho, Goodfire co-founder and CEO. “With the right interpretability tools, we can see how models actually work.” The tools Ho refers to are built around a concept called mechanistic interpretability, which aims to understand what goes on inside an AI model when it carries out a task by interpreting the model’s weights, activations, and attention patterns, and mapping its neurons and the pathways between them. Mechanistic interpretability tools span the gamut. One approach is mapping a model’s activations in response to controlled prompts, and matching those activation patterns to a set of human-understandable concepts. Another tack is tracking changes in model weights before and after a specific training run in order to spot and understand what changed. Yet another option is changing specific model weights or activations and observing how that affects the model’s output. With Silico, uSilico combines a broad range of these tools, and provides a layer of AI agents to help users understand their model. Users describe what they want to investigate about their AI model in plain language, asking things like ‘Find out when and why my model is hallucinating.’ The platform then autonomously builds an experimental plan involving a host of tasks that can be performed using the various interpretability tools and techniques at its disposal. It then sends out agents to perform these tasks in parallel. Completion of these subtasks should add up to an answer to the original prompt, or at least insights that can be inspected and built upon. Ho says: “In a sense, Silico is like a microscope to peer inside an AI model to understand which parts are responsible for what behavior, and even edit those parts directly.” Understanding Alzheimer’s and AI These tools have already been used to make some impressive advances in a host of fields. In medicine, for instance, Prima Mente, a UK-based AI company, worked with Goodfire to understand its Pleiades epigenetic foundation model. The model performed well at its task of detecting Alzheimer’s disease from blood samples, but the company didn’t know why. “We reverse-engineered Pleiades and found it was using DNA fragment-length patterns to make its predictions—a signal humans hadn’t used to detect Alzheimer’s before,” recalls Ho. In other words, the team had discovered that Pleiades was using a completely new biomarker for the disease. “As far as we know, it’s the first significant finding in the natural sciences discovered purely by reverse-engineering a foundation model,” Ho adds. Elsewhere, Silico is being used to explore deep questions surrounding AI. Cameron Berg, Founder and Director of Reciprocal Research (a New York nonprofit research organization he founded to explore methods of gauging AI cognition), says that Silico almost fell out of the sky at the right time for him and his research. “Silico has been really helpful for operationalizing my research agenda and executing on it way faster than I would have expected,” he says. “ I feel like I have basically become the PI and my research scientists and research engineers are AI systems.” Berg sees general access to Silico and tools like it leading to greater trust in AI’s ability to conduct research tasks, which will accelerate the scientific process across the board. But beyond scientific research, the widespread release of Silico could signal a shift in how AI innovators build, debug, and deploy their models. “I think it’s a mistake to not understand the most consequential technology of our time, particularly given the emergent behavior we’re seeing from increasingly capable AI agents,” says Ho. “If we truly understand how AI models think, instead of discovering and trying to correct their behavior retroactively, we can design them intentionally and shape how models behave to be safer and more reliable.”

Even the best AI models can suck at chess The launch of ChatGPT had an interesting effect on the online chess discourse. Chess has already long been conquered by machines. As early as 1996 a computer (IBM’s Deep Blue) was able to beat the human world champion, grandmaster Garry Kasparov, in a game watched by […]

<blockquote cite="https://pauldix.com/the-end-of-programming"><p>The fact that AI wrote 1M LOC and then refined it over the course of the next couple of months to produce a reliable piece of software that is currently running on millions of developer machines is absolutely mind blowing. And you can say, “well it’s not that impressive because they had an oracle to compare against, so it was simple to go from one language to another”, but I think that’s selling this entire thing short. If you can build a verification system and give proper direction, AI can produce a highly complex, highly sophisticated piece of software and it can continue to refine it until it just works.</p></blockquote> <p class="cite">&mdash; <a href="https://pauldix.com/the-end-of-programming">Paul Dix</a>, The end of programming</p> <p>Tags: <a href="https://simonwillison.net/tags/coding-agents">coding-agents</a>, <a href="https://simonwillison.net/tags/ai-assisted-programming">ai-assisted-programming</a>, <a href="https://simonwillison.net/tags/generative-ai">generative-ai</a>, <a href="https://simonwillison.net/tags/bun">bun</a>, <a href="https://simonwillison.net/tags/ai">ai</a>, <a href="https://simonwillison.net/tags/llms">llms</a></p>

IBM has released Granite 4.2, a family of open reasoning language models in 3B, 8B, and 30B sizes, all under Apache 2.0. Every model exposes a thinking / low-effort / non-thinking switch and native tool calling. The 8B and 30B additionally go through an agentic RL block that trains them to edit code, drive a terminal, and run web searches inside real sandboxed environments. The 30B reports 57.00 on SWE-Bench Verified and 29.24 on Terminal-Bench 2.1. The post IBM Releases Granite 4.2: Bringing Native Reasoning and Agentic RL to Open Enterprise Models appeared first on MarkTechPost.

Hey all, The goal is to earn on token margins for LLM calls when you build an AI-powered webapp. I proxy OpenAI and Anthropic calls so that when you deploy a site to a subdomain, your users token usage will be tracked.…

arXiv:2608.23831v1 Announce Type: new Abstract: While reinforcement learning (RL) allows generalist robot policies to continually improve during deployment, the large model size of modern generalist policies, such as VLAs, poses a fundamental obstacle to effective RL improvement. In particular, their severe inference latency---which can lead to pauses or jerky movements---can alter the effective environment dynamics and, if not correctly accounted for, break the Markov assumption that RL relies on, causing standard RL algorithms to fail completely. In this work, we introduce a latency-aware framework, Asynchronous RL with Intermediate Information (ARLI), that enables RL-based improvement of generalist policies under inference delays. Our framework builds on asynchronous inference approaches, which interleave action generation with execution to hide latency, and addresses its incompatibility with RL by providing a low-latency RL policy design that maximizes reactivity within the inference window through two contributions: state augmentations that restore near-Markovian structure by incorporating committed actions and a mid-inference observation. We evaluate our approach across simulated and real-world manipulation tasks, and find that it enables effective finetuning under inference delays where standard RL fails entirely, even matching or exceeding the performance of standard RL in idealized no-latency settings.

arXiv:2608.23752v1 Announce Type: new Abstract: The growing size of Convolutional Neural Networks has led to increasingly large and costly models. Knowledge Distillation (KD) addresses this by transferring knowledge from a large network (teacher) to a small one (student), also reducing the training data required. KD is traditionally applied only at the network's final output. However, its behaviour when applied at intermediate network layers has received little attention. This raises the question of whether intermediate block-wise KD, which provides supervision throughout the network, could offer an advantage under specific conditions, such as few instances per class, which is common in fine-grained datasets. This work proposes a student design based on simple, homogeneous blocks mirroring those of the teacher, distilling knowledge between corresponding blocks. Across eleven datasets, we show that on classic datasets, distilling only the last block is sufficient -- and often best--, whereas fine-grained, data-scarce settings benefit substantially from intermediate supervision, with even a single additional distillation point narrowing the gap considerably. We further study how this supervision should be guided, exploring configurations of varying granularity and informed by an explainability analysis based on attention maps, Centered Kernel Alignment, and Grad-CAM, alongside the impact of teacher and student fine-tuning strategies. This work shows that intermediate block-wise distillation, guided appropriately, is key to building compact data-efficient models without sacrificing accuracy.

arXiv:2608.23746v1 Announce Type: new Abstract: State space models, especially Visual State Space Duality (VSSD), have emerged as efficient linear-time alternatives to Transformers for dense visual tasks. However, we observe that VSSD compresses spatial context into a global aggregation that suppresses high-frequency responses, causing excessive boundary smoothing in remote sensing semantic segmentation. To address this, we propose CRISP, a calibration framework with two components. Its core, the Duality Calibration Operator (DCO), restores local contrast and boundary responses through residual injection and frequency calibration within the VSSD backbone, without altering its linear complexity. To retain the recovered detail, an Orthogonal Multi-Prototype (OMP) head assigns multiple orthogonally constrained prototypes per class to model large intra-class variance. Extensive experiments on Potsdam, Vaihingen, and LoveDA show that, with approximately 30M parameters, CRISP achieves consistent gains in mean F1 (mF) and mIoU while remaining competitive with state-of-the-art methods. Code is available at https://github.com/crazylifeha/CRISP.

arXiv:2608.23728v1 Announce Type: new Abstract: Satellite orbit prediction, which aims to forecast future orbital trajectories from historical observations, is important for collision warning and safe space operations. With advances in time-series forecasting, learning-based methods have emerged as a promising solution for satellite prediction. In orbital dynamics, a satellite state is typically described by position and velocity, where position characterizes trajectory geometry and velocity reflects its instantaneous direction and rate of change. However, most existing methods mainly focus on temporal dependencies within position sequences while rarely exploiting the intrinsic coupling between position and velocity, which is essential for modeling satellite motion. To this end, we propose OrbitNet, a velocity-aware representation learning method for accurate satellite orbit prediction. It lifts conventional position-sequence forecasting to a position-velocity coupled representation learning paradigm by exploiting relationships among satellite state variables. Specifically, we develop a velocity-coupled representation refinement strategy to enhance positional representations through cross-variable interactions between position and velocity. We further introduce orbital segment modeling, which partitions historical trajectories into temporal segments and performs segment-level temporal learning to capture local motion variations and long-range evolution patterns. Extensive experiments show that OrbitNet outperforms large time-series foundation models and representative general forecasting methods under both in-domain evaluation on Starlink and zero-shot evaluation across six unseen satellite constellations. We expect this work to encourage further exploration of satellite-aware representation learning for trajectory time-series forecasting.

arXiv:2608.23723v1 Announce Type: new Abstract: Few-shot anomaly detection (FSAD) has recently benefited from vision-language models such as CLIP, which enable anomaly de?tection by aligning visual features with text descriptions of normal and abnormal states. However, existing methods typically rely on static text prompts that are applied uniformly across the entire feature hierarchy and spatial dimensions. This rigid global-to-local matching fails to capture the highly localized and scale-dependent physical variations of industrial defects. To address this, we propose DriftAD, a FSAD framework built on three key modules. First, an Anomaly Signal Amplification (ASA) module enhances subtle defect signals through spatial and frequency branches before text-visual matching. Second, Visually-Guided Text Drift (VGTD) dynamically transforms frozen CLIP text embeddings, steering them into layer?wise, spatially-adaptive anomaly descriptors conditioned on local visual context at each encoder depth. Third, Drift-Guided Spatial Gating (DGSG) uses the drifted abnormal descriptor as a spatial probe to selectively enhance anomaly-relevant visual features. Addi?tionally, a drift separation loss prevents representational collapse of the drifted descriptors, and a gate supervision loss enforces spatially discriminative gating in DGSG. Extensive experiments on MVTec?AD and VisA demonstrate state-of-the-art performance across all 1-, 2-, and 4-shot settings on both image-level and pixel-level metrics. Code is available at https://github.com/wenyang001/DriftAD.

arXiv:2608.23664v1 Announce Type: new Abstract: Reward fine-tuning is becoming an important tool for adapting diffusion models to human preferences and task-specific objectives, but existing methods largely inherit policy-gradient machinery from large language models. Unlike autoregressive models, diffusion models do not provide tractable likelihoods for generated samples. As a result, current approaches either construct trajectory likelihoods from stochastic denoising transitions or approximate endpoint likelihoods with evidence lower bound, introducing additional computation and algorithmic complexity. We demonstrate that this likelihood-based machinery is not necessary for effective diffusion reward fine-tuning. We propose reward-based velocity matching (RVM), a simple trajectory-free update that acts directly on the velocity field. RVM reinforces directions associated with high-reward generations, suppresses those with low reward, and involves an optional anchor term controlling drift from a reference velocity. Notably, it provides a general framework that recovers recent fine-tuning methods, including RAM and DiffusionNFT, as special cases. Across various large-scale diffusion models reward fine-tuning tasks, RVM is competitive with or outperforms trajectory-based policy-gradient methods under substantially reduced training cost. We further find that, once the velocity update is simplified, the particular loss variant matters less than reward and anchor design. For video generation, standard preference rewards can favor visually clean but nearly static outputs; introducing a new dynamic-tracking reward that substantially improve motions while improving overall VBench performance. These results suggest that scalable reward fine-tuning for diffusion models is better posed in the native velocity representation than as likelihood-based policy optimization.

arXiv:2608.23634v1 Announce Type: new Abstract: Many few-shot adaptation methods for vision-language models classify with a convex combination of the zero-shot text prototype and the mean of the K labelled image features, with a single blending ratio routinely tuned on held-out labels, often on the test set itself. We ask what the family's own bias-variance justification invites: what is the right ratio, can it be estimated without validation data, and is finding it where the performance is? First, the ratio minimising prototype mean-squared error has a closed form whose support-set plug-in is exactly a positive-part James-Stein coefficient shrinking towards the text prototype. Across 4,800 cells (ten datasets, five backbones including SigLIP, five shot counts, five seeds, four prompt tiers) this theoretically optimal ratio is a reliable estimate of the wrong quantity: on the 950 primary-tier cells where it is defined it trails a test-set-oracle ratio by 8.5 points. It saturates near 1, discarding the text prior for a nearest-class-mean classifier, because 78% of the text-image prototype distance it treats as bias is a class-independent offset that the arg max largely cancels. We prove the mechanism and bound its share of the damage at 26% by a counterfactual. Second, leave-one-out on the support set alone sets a ratio landing within 0.9 points of the oracle blend, so it is estimable without validation data. Third, validation-free linear probes beat even the oracle-tuned blend: CLAP by +1.9 points and LP++ by +1.5 on average, and at K >= 4 all four validation-free baselines sit above the oracle, the linear probes by margins excluding zero. These results locate the ceiling in the model class, not the hyperparameter: the ratio can be set near-optimally for free, and it is still not where the performance is. Code, cached features, per-cell records: https://huggingface.co/datasets/Liangzhi-Li/clipbench-blending

arXiv:2608.23794v1 Announce Type: new Abstract: Mixture-of-Experts (MoE) scales language models by routing each input through a small set of independently parameterized experts. We show that copying this design into convolutional networks fails for a structural reason: parallel convolutional experts that read the same input channels learn nearly identical filters. We therefore move the expert axis from operator duplication to channel selection. We introduce Mixture of Channel Experts (MoCE), a structured sparse channel-mixing layer, inspired by MoE, that replaces pointwise (1x1) channel-reduction projections. In MoCE, an expert is a single output channel with a learned sparse support of k << C input channels. The selected channels are combined by a softmax whose temperature is predicted per input, so each expert can move between mean-like and max-like aggregation. A residual expert summarizes the unselected channels, and a load-balancing loss keeps channel coverage complete. MoCE replaces a dense projection whose cost is quadratic in C with a mechanism whose relative cost scales as k/C, and the predicted savings hold in measured wall-clock time. Across ResNet backbones on ImageNet-1K and CIFAR-100, transfer learning, EfficientViT, and a strong modern training recipe, MoCE matches or exceeds dense baselines and prior channel-selection methods while reducing MACs by 16.7% and end-to-end latency.

arXiv:2608.23782v1 Announce Type: new Abstract: Continual learning faces the persistent challenge of catastrophic forgetting, where sequential task updates degrade previously acquired knowledge. While prompt-based methods integrated with pre-trained models offer a compelling solution by freezing the backbone, they often rely on static, task-level prompting strategies that overlook fine-grained intra-task diversity. In this paper, we propose Gated Adaptive Prompting (GAP-Prompt), a novel method that introduces instance-level adaptability to the prompting process. GAP-Prompt consists of three synergistic modules: (1) instance-conditioned gating, which dynamically determines optimal prompt injection layers for each individual image; (2) dynamic knowledge fusion, which performs instance-aware aggregation of current and historical prompts, enabling knowledge integration across tasks; and (3) shared prompt distillation, which anchors foundational knowledge in early shared layers to mitigate forgetting. Extensive evaluations on CIFAR-100, ImageNet-R, and CUB-200 benchmarks demonstrate that GAP-Prompt consistently achieves state-of-the-art performance. Notably, on the fine-grained CUB-200 dataset, GAP-Prompt reaches 87.29% accuracy, approaching the joint training upper bound (88.00%) and outperforming existing methods by a significant margin.

arXiv:2608.23776v1 Announce Type: new Abstract: Understanding human skill is important for AI systems that collaborate with, coach, or assist people. Unlike typical latent variable estimation problems which rely on single observations, skill is a persistent, compositional, and behaviorally grounded construct that must be inferred from patterns over time. We introduce Skill Abstraction with Interpretable Latents (SAIL), a method for modeling human skill as an interpretable, multi-dimensional construct inferred from naturalistic behavior. Our approach produces a skill embedding that is robust to transient performance fluctuations and learns a transferable representation of human subskills. Furthermore, SAIL supports skill-informed behavior prediction that generalizes across a variety of in-domain contexts. We represent each individual with a persistent skill embedding that controls a blend between expert and novice bases and is trained using counterfactual subskill swaps for disentanglement. This design encourages representations that are both robust to performance variation and structured for interpretability. We demonstrate across racing and baseball that SAIL achieves strong predictive performance and consistently improves behaviorally grounded disentanglement over the evaluated baselines, while also improving downstream AI coaching performance.

arXiv:2608.23744v1 Announce Type: new Abstract: Split conformal prediction, not the pruning rule, supplies finite-sample marginal coverage once a pruned model is fixed independently of the conformal calibration split. We study the separate efficiency problem: can pruning preserve score geometry well enough to obtain smaller valid prediction sets? Calibration-Preserving Pruning (CPP) augments a base pruning score with nonconformity-gradient saliency and uses disjoint pruning, validation-selection, conformal-calibration, and test splits. Bounded score perturbations imply bounded conformal-quantile shifts and controlled set inflation, but do not make the generic coverage theorem CPP-specific. Final five-seed Qwen2.5-1.5B results at 50\% sparsity show the largest gains on large-label tasks. On DBpedia-14, CPP-SparseGPT reduces mean set size from \(10.1\) to \(8.6\) while changing accuracy from \(0.347\) to \(0.366\); CPP-Wanda reduces \(11.2\) to \(9.0\) with an accuracy trade-off from \(0.310\) to \(0.295\). Across 15 dataset--sparsity cells, CPP-SparseGPT produces smaller sets in 13 and higher accuracy in 11. Matched controls show that generic supervised gradients explain much of the gain: true-label CPP is not statistically resolved from matched Wanda+SNIP, whereas threshold-aware candidate-label CPP reaches \(7.8\) mean set size at explicit accuracy and offline-compute costs. RoBERTa-base and Llama-3-8B diagnostics support transfer, but our claims remain limited to reliability-sensitive classification.

arXiv:2608.23696v1 Announce Type: new Abstract: Despite their remarkable success in modeling complex data, generative models face a fundamental tradeoff. Global approaches can capture full structural coherence but suffer from high computational costs, while local models are efficient but often fail to reproduce long-range correlations and global coherence. The renormalization group (RG) bridges this gap by seamlessly connecting spatial structures across different length scales, retaining quasi-local descriptions at each step while preserving long-range correlations. We introduce renormalization group flow matching (RGFM), a generative framework that systematically structures data generation across different spatial scales. By using an exact RG flow as the probability path, RGFM progressively generates data from long- to short-wavelength structures. To reconcile scalability with global structure, we exploit two key properties of the RG: quasi-locality and scale separation. We rigorously show that the RGFM probability flow can be accurately approximated by local velocity fields acting over a spatial range $O(\Lambda^{-1}[\ln L+\ln(1/\varepsilon)])$ for RG wavenumber scale $\Lambda$, linear system size $L$, and prescribed error tolerance $\varepsilon$. This property enables local generative modeling with patches of size $O(\ln L)$ and a computational cost that scales nearly linearly with the system volume. We numerically demonstrate that local RGFM reproduces long-range correlations far beyond its receptive field in representative one-dimensional distributions, while conventional local flow matching exhibits substantial errors at long distances. On FFHQ images, RGFM yields far more coherent and higher-quality samples than local flow matching at 64x64 and 256x256. Our results establish RG-guided probability flows as a promising route toward scalable generative modeling that captures long-range structure using only local computation.

arXiv:2608.23660v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly used to provide prior causal knowledge for structural causal discovery, yet whether their direct-edge judgments and confidence can be trusted remains unclear. We systematically evaluate 12 instruction-tuned open-weight models across six benchmark causal graphs, five prompting strategies, and four confidence sources: verbalized, logit-based, cross-prompt agreement, and cross-model agreement. Under our language-only pairwise protocol, our evaluation yields three key findings. (i) LLM-based causal judgments are strongly recall-dominant: models predict overly dense graphs with many false-positive edges, while prompting mainly shifts the precision-recall trade-off rather than resolving overprediction. Gains from model scale diminish on the largest graphs and do not eliminate miscalibration. (ii) LLMs often capture causal relatedness without reliably identifying directness or orientation. Relative to published reference graphs, models misclassify 40.0% of indirect and 36.0% of reversed non-edges as direct edges, versus 28.2% of other non-edges. Moreover, 80.8% and 84.6% of these false positives receive verbalized confidence of at least 80%, revealing substantial overconfidence in structurally incorrect predictions. (iii) Conventional confidence estimates are unreliable, whereas agreement offers a more promising signal. Logit-based confidence frequently collapses near 1.0 regardless of correctness, while cross-prompt and cross-model agreement achieve better mean calibration and discrimination, though their advantages are not statistically significant after Holm correction. A benchmark-familiarity audit further identifies potential familiarity in five model-dataset pairs, all involving AsiaM. Overall, our results suggest LLMs are better viewed as sources of externally validated soft causal priors than as direct evidence of causal structure.

arXiv:2608.23573v1 Announce Type: new Abstract: A trained transformer's weight magnitudes can be summarized by a two-parameter Weibull distribution whose shape $k \approx 1.2$ is stable across layers and models, so the scale $\lambda$ carries most training-induced movement. What corpus property sets how much $\lambda$ grows? Using the bigram conditional entropy $D = H(\text{next} \mid \text{prev})$, a training-free statistic computed before training, we find across controlled corruption families a learning-rate-conditioned law, $\lambda^2 - \lambda_0^2 = C_0(\eta) + C_1(\eta)(H_r - D)^{0.59}$, where $H_r$ is a matched-budget shuffle baseline. The convex exponent is inherited from an independently measured data-side saturation relation rather than fitted directly to the growth curve. After removing the two per-$\eta$ coefficients, 23 runs spanning an order of magnitude in learning rate collapse onto $(H_r - D)^{0.59}$ with unit slope ($R^2 = 0.941$; direct per-$\eta$ fits are weaker, $R^2 \approx 0.82$). Because $D$ is computed before training, the law is a forward predictor: an end-to-end self-validation recovers held-out within-family weight growth with 5.7% relative error. The readout holds at model and per-layer resolutions and across two tested architectures, with the functional form preserved and only the coefficients changing. It also marks its boundary: cross-corpus prediction over-predicts code, implicating redundancy as a second axis of a broader $\Phi(D,R,A,H)$ data-to-weight framework.

arXiv:2608.23571v1 Announce Type: new Abstract: Equivariant message-passing networks are the standard model for molecular property and interatomic-potential prediction, and recent work predicts the electronic Hamiltonian itself in an E(3)-equivariant way. Separately, topological deep learning has extended graph networks to cellular sheaves. Our central observation is structural: in a localized atomic-orbital basis, the molecular single-particle Hamiltonian, after a constant shift that makes it positive semidefinite, is the Laplacian of a cellular sheaf on a regular cell complex built from the molecule. Making the restriction maps O(3)-steerable two-center kernels from bond geometry recovers the Slater-Koster form as a special case and yields an E(3)- and permutation-equivariant operator. Three consequences follow. First, the zeroth sheaf cohomology H^0 = ker L is a topological invariant equal to the non-bonding (zero-mode) orbitals, recovering the classical alternant non-bonding-orbital count as a lower bound. Second, the Hodge 1-Laplacian lets higher cells (rings) carry cycle and delocalization information through H^1. Third, the model strictly generalizes E(3)-equivariant message-passing networks and CW networks, and inherits the anti-oversmoothing of non-trivial sheaf diffusion. We prove equivariance, expressivity, and cohomological-correspondence results for the Equivariant Cellular Sheaf Networks, and validate them numerically: the Hamiltonian-to-sheaf embedding is exact to machine precision, the cohomology dimension reproduces non-bonding-orbital counts across eleven conjugated molecules, the sheaf Laplacian is O(3)-equivariant to machine precision, and the equivariant model attains lower error and rotation generalization on a directional electronic target. Our contribution is this sheaf-theoretic formalization and its invariants, not equivariant Hamiltonian prediction itself.

arXiv:2608.23640v1 Announce Type: new Abstract: When a large language model (LLM) is asked to write a person's life, how much of what it writes actually happened? We present a scene-level case-study audit - the first quantified audit of LLM-generated autobiography against a subject-specific ground-truth corpus that we are aware of, based on an unsystematic literature search. The subject and the author of this paper are the same person: a 366-day "page-a-day" book of first-person anecdotal entries was drafted with a conversational LLM whose documented inputs were a template, two exemplar days, and each day's quote - not her corpus - and every day was subsequently audited at the anecdote-scene level against an independent verification corpus using a four-level rubric fixed before analysis. We define the verification-failure rate as the share of days not rated VERIFIED (scene positively corroborated): 354 of 366 days fail, 96.7% (Wilson 95% CI 94.4-98.1%). Only 12 days contain a corroborated scene; 19 days (5.2%) assert claims actively contradicted by the record; the dominant failure mode is grounded drift - real people, employers, and settings inside invented scenes - though its measured share varies across raters. Independent re-rating replicates the headline (no evidence the original rate was inflated) while showing that the four-way taxonomy has only fair-to-moderate reliability. Regenerating the same days with current named models reproduces 100% verification failure under the same inputs; grounding generation in the subject's corpus significantly improves the verification rate while leaving substantial residual failure (83.3%). We contribute the measurement, a reusable audit instrument whose WEAK/UNVERIFIED boundary we show to be unreliable, and a grounding remedy with quantified effect.

arXiv:2608.23632v1 Announce Type: new Abstract: Process supervision has improved mathematical reasoning, where intermediate steps are naturally expressed as chains of thought. In code generation, however, process supervision remains underexplored because there is no standard notion of a step. Supervision can target lines, reasoning traces, or program states, making it unclear what to label and optimize. We propose STEP-KTODER, a framework for code preference optimization that defines steps as module-level functions in decomposed multi-function programs and assigns binary correctness labels via automatically generated unit tests. Our method provides a code-specific instantiation of stepwise KTO, combining function-level process supervision with outcome-level feedback on the full program. We evaluate on HumanEval(+), MBPP(+), BigCodeBench, and LiveCodeBench, showing that STEP-KTODER improves over outcome-only KTO and DPO. Further analysis shows that execution-based labels are essential: LLM-as-a-judge annotations systematically over-predict function failures, corrupt positive step labels, and degrade downstream preference optimization. Code is available at: https://github.com/inechnech/STEP-KTODER.

arXiv:2608.23631v1 Announce Type: new Abstract: Multi-objective materials discovery with LLM agents is often limited not only by how many candidates can be proposed, but by how effectively each costly property evaluation informs the next search step. Existing agents mainly store evaluated candidates and their scores, so they know which materials succeeded but not which executable edits caused useful property changes. This makes local refinement difficult when objectives compete and an edit that improves one property may damage another. We propose TRACE, a transition-aware residual control framework that treats evaluated edits as the basic unit of feedback. TRACE records each local refinement as a parent-edit-child transition with observed property deltas, aggregates transition evidence to estimate reusable edit effects, and ranks future edits by their predicted ability to reduce the current candidate's remaining constraint violations while avoiding damage to already satisfied objectives. In a controlled same-backbone comparison, TRACE improves over LLEMA, the state-of-the-art LLM-agent baseline, raising macro-average hit rate from 18.13\% to 25.96\%.

arXiv:2608.23626v1 Announce Type: new Abstract: Foundation models for astronomy are trained on survey pixels together with the catalogue products derived from those pixels. Those catalogues are incomplete at a measurable rate, and a model trained on both inherits that incompleteness as a systematic. We audit AION-1, a 39-modality transformer trained on more than 200 million objects, using causal interventions on its inputs. Holding the image tokens byte-identical and editing only the survey segmentation map changes every quantity the model reports -- flux, size, ellipticity, redshift -- by 110-4400 times a matched placebo. The mechanism is detection gating, presence at the field centre (r = 0.47), not the light the mask encloses (r = 0.30); across 322 real blends the model ignores how the pipeline partitioned the light (R = -0.006). Nor is the preference specific to that channel: contradicted catalogue photometry leaves the model nine times worse than supplying no metadata at all. The Legacy Survey pipeline leaves 3.68% of targets with no segment covering their position. Propagating that rate, with a miss represented by the fields the pipeline actually returns, shifts tomographic mean redshifts by a median 0.71 times the LSST DESC requirement over 40 assignments and exceeds it in 12; observed positional errors take the worst bin to 8.3 times. Drawing the misses by their measured magnitude dependence rather than uniformly does not change it. Spectroscopy removes the effect, withholding the detection channel removes it at no measurable cost, and the effect grows with model scale. Two further limits lie in the tokeniser: its image codec resolves 28 effective states on source patches against 934 for the spectrum codec, and the redshift readout is quantisation-limited. Sparse dictionaries are unreliable causal handles: across 15, recovery spans 26-75% and moves up to 18 points on the seed alone.

arXiv:2608.23622v1 Announce Type: new Abstract: Large language models (LLMs) have shown strong capabilities in reasoning, planning, and tool use, but many scientific and engineering tasks require more than plausible text and code generation. They require understanding how a system responds to intervention, which in practice depends on controlled experimentation. In this work, we propose a multi-agent framework that enables LLM agents to conduct controlled experiments with scientific simulation models for pharmaceutical process design. Given a user query and a baseline configuration, the system constructs a structured task representation, designs experiments, executes comparative simulation, interprets the resulting outcomes, and synthesizes evidence-based recommendations for process parameter optimization. By coupling language models with high-fidelity simulation models in an interactive agent framework, the proposed system supports reasoning through intervention, comparison, and observation. As a result, it produces more specific and actionable outputs than language-only reasoning. In an industrial application setting, this advantage is reflected in higher output specificity as well as improved user-rated correctness and helpfulness. Ablation studies and visualized case analyses further demonstrate the effectiveness and practical utility of simulation-integrated experimental reasoning.

arXiv:2608.23569v1 Announce Type: new Abstract: State-of-the-art Natural Language to SQL (NL2SQL) models report execution accuracy exceeding 89 percent on established benchmarks such as Spider and BIRD. However, these benchmarks rely on simplified academic schemas and open-source SQL dialects that do not reflect the complexity of enterprise database environments. We introduce ESQ-Bench, an Oracle-first NL2SQL benchmark with systematic complexity tiers and silent-divergence evaluation across three enterprise schema complexity tiers. We constructed and released six populated schemas (465 tables, 164,682 rows, zero empty tables) with identical seed data on Oracle, PostgreSQL, MySQL, and SQL Server, a four-metric evaluation harness (EM, EX, SR, SD), and 550 gold-validated question-query pairs (Tier-1: 95; Tier-2: 228; Tier-3: 227). Schema-linked prompting with GPT-4o shows monotonic execution-match degradation across tiers: 79.8, 60.3, and 57.2 percent EX on executed queries (June 2026), versus 75.6, 80.4, and 95.8 percent on an earlier 142-question pilot slice. EM stays below 7 percent tier-wide; operational silent-divergence reaches 73 to 99 percent among EX-passing queries. Failure analysis shows wrong-result semantics dominate at higher tiers. Claude Sonnet 4.6 with schema-linked prompts reaches 87.4, 74.9, and 68.7 percent EX (executed queries), exceeding GPT-4o schema-linked on every tier. GPT-4o zero-shot EX on executed queries (78.7, 73.5, and 77.8 percent) inverts schema-linked at Tiers 2 to 3 due to lower execution rates and survivor bias in the zero-shot versus schema-linked analysis. Local Llama 3.2 schema-linked reaches only 13.3 percent bank-wide EX (73 out of 550), underscoring the gap between closed API models and open-weight baselines on enterprise Oracle schemas.

arXiv:2608.23568v1 Announce Type: new Abstract: Memory and RAG evaluations often treat the answering model's input as an implementation detail, even though systems may render the same history as a memory entry, summary, typed record, or raw excerpt. We introduce RENDER, a benchmark control that fixes the conversation while varying the reader-facing artifact. RENDER combines a five-level packet ladder, localizing when answer-bearing content enters the input, with deterministic templates approximating ChatGPT-style entries, LangChain summaries, MemGPT-style typed records, and raw conversation. On 500 LongMemEval questions and nine models, matched-budget resolved packets beat recency-truncated raw dialogue by 42.4-72.6 points. In deployed-style templates, best-worst spread is 24.6-48.8 points per model; under the primary scorer, ChatGPT-style entries have higher point estimates than raw conversation on 7 of 9 models. Judge rescoring preserves the positive aggregate effect, but model-specific significance is mixed. Three models scoring 0 percent on formal ledger packets answer the same facts from natural-language entries at 45.4-53.4 percent. The effect persists under retrieval noise and transfers to HotpotQA, suggesting that memory/RAG evaluations should report or control the reader-facing artifact.

Maria Sukhareva Aug 25, 2026 ∙ Paid Hype-free executive briefing on last week’s critical AI developments, complete with ready-to-present slides for your team. If you are a paid subscriber, you can listen to the radar in…

Model cards report quality under server-class, full-precision conditions. Those numbers rarely predict how the same model behaves on a phone. This week, Liquid AI released Pipette. It is an open-source platform for benchmarking foundation models on edge devices, built in partnership with Artificial Analysis as an independent methodology validator. Pipette treats on-device behavior as a […] The post Liquid AI Open-Sources Pipette: A Reproducible Benchmarking Suite That Measures On-Device Models, Quantization, Runtime and Hardware Together appeared first on MarkTechPost.

These are the lessons we learned evaluating LLMs for real-world secret scanning. The post How to evaluate LLMs before production appeared first on The GitHub Blog.

Learn how to fine-tune and evaluate LLMs with LangSmith for dataset management. Complete guide covers LLaMA2 and GPT-3.5 fine-tuning with practical examples.

Build better Q&A systems for CSV data using LangChain agents, retrieval, and LLM evaluation. Includes benchmarks, debugging insights, and open-source code.

LangChain secures $10M seed round from Benchmark to empower developers building AI apps with our open-source framework for data-aware, agentic LLMs.

Scale retrieval apps to production with LangChain's Airbyte integration. Automate data ingestion with scheduling, text splitting, and 50+ embeddings.

Build AI-powered data experiences with Cube's semantic layer and LangChain. Prevent hallucinations, query in natural language, create conversational interfaces.

Automate web research with LangChain's retriever. Run parallel searches, scrape pages, and synthesize information with LLMs—locally or in the cloud.

On Tuesday, IBM launched the latest family of its open-weight Granite large language models (LLMs). Weighing in at 3 billion, The post IBM’s new Granite 4.2 models add reasoning and stay dense appeared first on The New Stack.

Query SQL databases using natural language with LLMs. Learn techniques to reduce hallucinations and build reliable text-to-SQL solutions with LangChain.

Build multi-modal RAG apps for slide decks using GPT-4V. Compare approaches, evaluate with benchmarks, and deploy with LangChain templates for visual Q&A.

Learn context engineering: building dynamic systems that provide LLMs the right information, tools, and format to reliably accomplish tasks.

Training and serving frontier models is now a networking problem as much as a compute problem. Collective operations like all-reduce and all-to-all synchronize thousands of accelerators during training, and the slowest transfer sets the pace for the entire job. Even small amounts of network friction directly strand significant compute capacity. This week, Meta introduced MetaRoCE. […] The post Meta AI Introduces MetaRoCE: A Clean-Sheet RDMA Transport Built for AI-Scale Ethernet appeared first on MarkTechPost.

AI/LLM Usage Becoming A "Denial of Service Attack" On Open-Source Project Maintainers The latest criticism of the increased burden placed on open-source software project maintainers caused by AI / LLM agents is around t…

Nvidia Corp. today announced the release of Jetson Orin Nano 2, a robotics computer “brain” for running artificial intelligence and frontier-level models at the edge. In the past months, foundational AI models have grown smaller and more efficient, adding numerous capabilities alongside language understanding, computer vision and audio processing. As more AI models compress in […] The post Nvidia doubles compute for entry-level edge robotics with Jetson Orin Nano 2 appeared first on SiliconANGLE.

The French AI lab extends its push for regional control of AI from Europe to the Middle East.

OpenAI says its new AI chip, Jalapeño, completes tasks more efficiently and returns responses faster than other AI systems, according to a blog post published on Tuesday. During a briefing with reporters, OpenAI hardware vice president Richard Ho said Jalapeño offers the "best of both worlds" with lower latency and higher throughput, as AI systems typically "have to make a trade-off between the two." First introduced in June, Jalapeño is an Application-Specific Integrated Circuit (ASIC) made in partnership with Broadcom. It's designed for AI inference - the process of running a trained AI model to complete a task or deploy an agent. To mea … Read the full story at The Verge.

Running an AI model locally, Perplexity's new Portable Computer can deliver faster performance, tighter security, and lower costs. But it has some strict requirements.

Mathematicians are raising concerns that the technology could kill their profession. But they still have abilities AI doesn’t Earlier this month, about 40 top mathematicians gathered at OpenAI’s offices to discuss the future of their profession. The meeting was off-the-record, but if recent articles by mathematicians are any guide, it was mostly pretty glum. People fear for their jobs, their careers and the work they love. We think the contrary view is more likely, at least in the short-term. AI models are nowhere near as capable as experienced academic mathematicians. Continue reading...

--> AI Personality | Abby Steele top of page AI Personality Abby Steele Let's play Chess! As Featured On bottom of page

Xiaomi reveals AI Cube prototype: Multi-chip local LLM powerhouse takes aim at Apple's Macs Xiaomi has showcased its AI Cube prototype–an on-premise local computing appliance combining three in-house made XRING processo…

Artificial Analysis DeepSeek • DeepSeek V4 Flash 0731 • Proprietary model • Released August 2026 DeepSeek V4 Flash Vision (Reasoning, Max Effort) Intelligence, Performance & Price Analysis API Provider Benchmarks Model…