Daily Papers

Software testing is a critical, yet resource-intensive phase of the software development lifecycle. Over the years, various automated tools have been developed to aid in this process. Search-based approaches typically achieve high coverage but produce tests with low readability, whereas large language model (LLM)-based methods generate more human-readable tests but often suffer from low coverage and compilability. While the majority of research efforts have focused on improving test coverage and readability, little attention has been paid to enhancing the robustness of bug detection, particularly in exposing corner cases and vulnerable execution paths. To address this gap, we propose AdverTest, a novel adversarial framework for LLM-powered test case generation. AdverTest comprises two interacting agents: a test case generation agent (T) and a mutant generation agent (M). These agents engage in an adversarial loop, where M persistently creates new mutants "hacking" the blind spots of T's current test suite, while T iteratively refines its test cases to "kill" the challenging mutants produced by M. This interaction loop is guided by both coverage and mutation scores, enabling the system to co-evolve toward both high test coverage and bug detection capability. Experimental results in the Defects4J dataset show that our approach improves fault detection rates by 8.56% over the best existing LLM-based methods and by 63.30% over EvoSuite, while also improving line and branch coverage.

Large Language Models (LLMs) have significantly advanced automated test generation, yet existing methods often rely on ground-truth code for verification, risking bug propagation and limiting applicability in test-driven development. We present ConVerTest, a novel two-stage pipeline for synthesizing reliable tests without requiring prior code implementations. ConVerTest integrates three core strategies: (i) Self-Consistency(SC) to generate convergent test cases via majority voting; (ii) Chain-of-Verification (CoVe) for iterative, reasoning-guided code refinement; and (iii) a Dual Execution Agreement to crossvalidate code and tests through consensus. Experiments on BIGCODEBENCH and LESS BASIC PYTHON PROBLEMS (LBPP) benchmarks demonstrate that ConVerTest improves test validity, line coverage, and mutation scores by up to 39%, 28%, and 18% respectively over baselines. Our findings highlight ConVerTest as a robust solution for mitigating hallucinations and enhancing the reliability of autonomous software testing agents.

Coding agents are increasingly used to create large software projects. While code-generation benchmarks have scaled accordingly to full applications, test-generation benchmarks remain limited to individual functions or bug fixes. In particular, generating complete test suites for complex software from scratch remains unexplored. To address this gap, we introduce TestSuiteBench, a benchmark of 165 open-source applications, each with a structured and detailed specification of intended behavior, a reference implementation, a reference test suite, and on average 74 mutants. Each mutant is generated from a single requirement of the specification and refined to isolate that behavior: 73\% are killed by exactly one reference test and, on average, by only 0.2\% of the remaining tests. We find that even frontier coding agents struggle to generate strong test suites given our specifications, achieving mutation scores below 60\%, despite their suites being largely valid and 88\% of their tests passing on the reference implementation. We observe that generated test suites often lack relevant input combinations and are smaller than reference test suites. Meanwhile, implementations generated for the same specifications pass 82\% of the reference tests. These results show agents can build software they cannot verify, a gap that directly limits fully autonomous software development.

Generative AI is gaining increasing attention in software engineering, where testing remains an indispensable reliability mechanism. According to the widely adopted testing pyramid, unit tests constitute the majority of test cases and are often schematic, requiring minimal domain expertise. Automatically generating such tests under the supervision of software engineers can significantly enhance productivity during the development phase of the software lifecycle. This paper investigates the impact of code context and prompting strategies on the quality and adequacy of unit tests generated by various large language models (LLMs) across several families. The results show that including docstrings notably improves code adequacy, while further extending context to the full implementation yields definitely smaller gains. Notably, the chain-of-thought prompting strategy -- applied even to 'reasoning' models -- achieves the best results, with up to 96.3\% branch coverage, a 57\% average mutation score, and near-perfect compilation success rate. Among the evaluated models, M5 (Gemini 2.5 Pro) demonstrated superior performance in both mutation score and branch coverage being still in top in terms of compilation success rate. All the code and resulting test suites are publicly available at https://github.com/peetery/LLM-analysis.

We present Test-Driven AI Agent Definition (TDAD), a methodology that treats agent prompts as compiled artifacts: engineers provide behavioral specifications, a coding agent converts them into executable tests, and a second coding agent iteratively refines the prompt until tests pass. Deploying tool-using LLM agents in production requires measurable behavioral compliance that current development practices cannot provide. Small prompt changes cause silent regressions, tool misuse goes undetected, and policy violations emerge only after deployment. To mitigate specification gaming, TDAD introduces three mechanisms: (1) visible/hidden test splits that withhold evaluation tests during compilation, (2) semantic mutation testing via a post-compilation agent that generates plausible faulty prompt variants, with the harness measuring whether the test suite detects them, and (3) spec evolution scenarios that quantify regression safety when requirements change. We evaluate TDAD on SpecSuite-Core, a benchmark of four deeply-specified agents spanning policy compliance, grounded analytics, runbook adherence, and deterministic enforcement. Across 24 independent trials, TDAD achieves 92% v1 compilation success with 97% mean hidden pass rate; evolved specifications compile at 58%, with most failed runs passing all visible tests except 1-2, and show 86-100% mutation scores, 78% v2 hidden pass rate, and 97% regression safety scores. The implementation is available as an open benchmark at https://github.com/f-labs-io/tdad-paper-code.

Familial amyloid-β (Aβ) variants alter aggregation kinetics, but their interaction with structural brain connectivity remains incompletely understood. We developed a mutation-aware mechanistic model coupling a coarse-grained monomer--oligomer--fibril aggregation--fragmentation system to graph diffusion on the 540-node Budapest Reference Connectome component. Experimental Aβ_{42} nucleation scores scaled primary nucleation rates for seven variants relative to wild type. Robustness was examined using mutation-score uncertainty, alternative kinetic mappings, global sensitivity analysis, seed and edge-weight perturbations, degree-preserving randomized connectomes, spatial propagation analysis, synthetic ABC-SMC parameter recovery, posterior prediction, and Chemical Langevin simulations. E22G showed the earliest threshold crossing and greatest cumulative oligomer burden, whereas A2V was delayed under the selected mapping. Mutation rankings persisted across tested network perturbations, although regional burden patterns depended on topology. Connectome distance from seed regions was associated with later oligomer arrival (Spearman ρapprox0.92). Under inferred parameter uncertainty, the mean timing-rank correlation was 0.990 and cumulative-burden ordering was preserved in every posterior draw, while exact peak-amplitude ordering was less stable. Stochastic ensemble medians retained the deterministic ordering despite overlap among trajectories. Within this proof-of-concept framework, mutation-dependent kinetics primarily influence aggregation timing and cumulative burden, while connectivity shapes spatial propagation. Nominal background rates, model-time units, and synthetic parameter recovery limit interpretation to mechanistic comparisons rather than clinically calibrated prediction.

Mutation testing is an effective approach to evaluate and strengthen software test suites, but its adoption is currently limited by the mutants' execution computational cost. Several strategies have been proposed to reduce this cost (a.k.a. mutation cost reduction strategies), however none of them has proven to be effective for all scenarios since they often need an ad-hoc manual selection and configuration depending on the software under test (SUT). In this paper, we propose a novel multi-objective evolutionary hyper-heuristic approach, dubbed Sentinel, to automate the generation of optimal cost reduction strategies for every new SUT. We evaluate Sentinel by carrying out a thorough empirical study involving 40 releases of 10 open-source real-world software systems and both baseline and state-of-the-art strategies as a benchmark. We execute a total of 4,800 experiments, and evaluate their results with both quality indicators and statistical significance tests, following the most recent best practice in the literature. The results show that strategies generated by Sentinel outperform the baseline strategies in 95% of the cases always with large effect sizes. They also obtain statistically significantly better results than state-of-the-art strategies in 88% of the cases, with large effect sizes for 95% of them. Also, our study reveals that the mutation strategies generated by Sentinel for a given software version can be used without any loss in quality for subsequently developed versions in 95% of the cases. These results show that Sentinel is able to automatically generate mutation strategies that reduce mutation testing cost without affecting its testing effectiveness (i.e. mutation score), thus taking off from the tester's shoulders the burden of manually selecting and configuring strategies for each SUT.

One of the critical phases in software development is software testing. Testing helps with identifying potential bugs and reducing maintenance costs. The goal of automated test generation tools is to ease the development of tests by suggesting efficient bug-revealing tests. Recently, researchers have leveraged Large Language Models (LLMs) of code to generate unit tests. While the code coverage of generated tests was usually assessed, the literature has acknowledged that the coverage is weakly correlated with the efficiency of tests in bug detection. To improve over this limitation, in this paper, we introduce MuTAP for improving the effectiveness of test cases generated by LLMs in terms of revealing bugs by leveraging mutation testing. Our goal is achieved by augmenting prompts with surviving mutants, as those mutants highlight the limitations of test cases in detecting bugs. MuTAP is capable of generating effective test cases in the absence of natural language descriptions of the Program Under Test (PUTs). We employ different LLMs within MuTAP and evaluate their performance on different benchmarks. Our results show that our proposed method is able to detect up to 28% more faulty human-written code snippets. Among these, 17% remained undetected by both the current state-of-the-art fully automated test generation tool (i.e., Pynguin) and zero-shot/few-shot learning approaches on LLMs. Furthermore, MuTAP achieves a Mutation Score (MS) of 93.57% on synthetic buggy code, outperforming all other approaches in our evaluation. Our findings suggest that although LLMs can serve as a useful tool to generate test cases, they require specific post-processing steps to enhance the effectiveness of the generated test cases which may suffer from syntactic or functional errors and may be ineffective in detecting certain types of bugs and testing corner cases PUTs.

Recent work pairs LLMs with evolutionary search to iteratively generate, modify, and select code using task-specific feedback. These systems have produced strong results in mathematical discovery and algorithm design, yet a fundamental question remains: what do they actually evolve? Progress is typically summarized by the best score a run reaches under a task-specific evaluator, but that score can reflect several different mechanisms: new algorithmic structure, re-tuning an existing strategy, recombining ideas already in the model's internal knowledge, or overfitting to the evaluator. Distinguishing these mechanisms requires inspecting the search process itself, not only its final outcome. We introduce EvoTrace, a dataset of evolutionary coding traces spanning four evolutionary frameworks, reasoning and non-reasoning models, and 16 tasks across mathematics and algorithm design. To analyze these traces, we develop EvoReplay, a replay-based methodology that reconstructs the local search states behind high-scoring solutions and tests controlled interventions, including adjusting constants, removing program components and substituting models or prompting contexts. We annotate every code edit in EvoTrace with one of nine recurring edit types using an LLM-as-judge pipeline validated against blind human re-annotation. Across EvoTrace, most score gains come from a small subset of these edit types. We further find a deterministic cycling pattern: about 30% of code lines added during search are byte-identical re-introductions of previously-deleted lines, present throughout nearly every run. These results show that benchmark gains in evolutionary coding agents can arise from qualitatively different mechanisms, only some of which correspond to new algorithmic structure. EvoTrace enables more diagnostic evaluation of evolutionary coding agents beyond final benchmark scores.

In this paper we show that reporting a single performance score is insufficient to compare non-deterministic approaches. We demonstrate for common sequence tagging tasks that the seed value for the random number generator can result in statistically significant (p < 10^-4) differences for state-of-the-art systems. For two recent systems for NER, we observe an absolute difference of one percentage point F1-score depending on the selected seed value, making these systems perceived either as state-of-the-art or mediocre. Instead of publishing and reporting single performance scores, we propose to compare score distributions based on multiple executions. Based on the evaluation of 50.000 LSTM-networks for five sequence tagging tasks, we present network architectures that produce both superior performance as well as are more stable with respect to the remaining hyperparameters.

The ability to accurately model the fitness landscape of protein sequences is critical to a wide range of applications, from quantifying the effects of human variants on disease likelihood, to predicting immune-escape mutations in viruses and designing novel biotherapeutic proteins. Deep generative models of protein sequences trained on multiple sequence alignments have been the most successful approaches so far to address these tasks. The performance of these methods is however contingent on the availability of sufficiently deep and diverse alignments for reliable training. Their potential scope is thus limited by the fact many protein families are hard, if not impossible, to align. Large language models trained on massive quantities of non-aligned protein sequences from diverse families address these problems and show potential to eventually bridge the performance gap. We introduce Tranception, a novel transformer architecture leveraging autoregressive predictions and retrieval of homologous sequences at inference to achieve state-of-the-art fitness prediction performance. Given its markedly higher performance on multiple mutants, robustness to shallow alignments and ability to score indels, our approach offers significant gain of scope over existing approaches. To enable more rigorous model testing across a broader range of protein families, we develop ProteinGym -- an extensive set of multiplexed assays of variant effects, substantially increasing both the number and diversity of assays compared to existing benchmarks.

Benchmarks for evaluating large language models on register-transfer-level (RTL) hardware design have proliferated rapidly, yet none reports having applied mutation testing, an established hardware-verification technique for quantifying testbench quality, to ask whether its own testbenches are trustworthy. A testbench that never fails is not evidence of a correct design; it may simply never stimulate the logic that is actually broken. We introduce GateTruth, a mutation-testing engine and methodology for auditing RTL benchmark testbench rigor: inject a deterministic, seeded set of semantic mutants into a reference design and measure what fraction the testbench catches. We validate the methodology against our own 68-task, dual-track reference suite -- 60 specification-to-RTL generation tasks and 8 agentic-repair tasks, scored through a pinned, deterministic synthesis-to-timing flow with correctness enforced as a strict gate -- certifying that 46 of 60 Track A testbenches kill at least 95% of injected mutants under sequential, reproducible execution; we disclose why the other 14 do not, including a Goodhart effect on testbenches revised to pass this gate. We then point the same engine, unmodified, at RTLLM v2.0, a widely adopted external benchmark: of 46 auditable designs, 72% fall below the 95% floor our own suite is held to, and three score 0% outright. A comparable audit of NVIDIA's CVDP benchmark is structurally impossible: its public release withholds reference solutions, removing the golden RTL mutation testing requires. Auditing our own instrument also surfaced a second finding: an initially uniform 4096-token output cap silently truncated three of seven evaluated models, and re-running at 16,384 tokens moved one model from fifth place to first. We argue mutation-kill certification should become a standard reporting requirement for RTL-generation benchmarks generally.

In today's era, where large language models (LLMs) are integrated into numerous real-world applications, ensuring their safety and robustness is crucial for responsible AI usage. Automated red-teaming methods play a key role in this process by generating adversarial attacks to identify and mitigate potential vulnerabilities in these models. However, existing methods often struggle with slow performance, limited categorical diversity, and high resource demands. While Rainbow Teaming, a recent approach, addresses the diversity challenge by framing adversarial prompt generation as a quality-diversity search, it remains slow and requires a large fine-tuned mutator for optimal performance. To overcome these limitations, we propose Ferret, a novel approach that builds upon Rainbow Teaming by generating multiple adversarial prompt mutations per iteration and using a scoring function to rank and select the most effective adversarial prompt. We explore various scoring functions, including reward models, Llama Guard, and LLM-as-a-judge, to rank adversarial mutations based on their potential harm to improve the efficiency of the search for harmful mutations. Our results demonstrate that Ferret, utilizing a reward model as a scoring function, improves the overall attack success rate (ASR) to 95%, which is 46% higher than Rainbow Teaming. Additionally, Ferret reduces the time needed to achieve a 90% ASR by 15.2% compared to the baseline and generates adversarial prompts that are transferable i.e. effective on other LLMs of larger size. Our codes are available at https://github.com/declare-lab/ferret.

With the growing reliance on automated code completion tools in software development, the need for robust evaluation benchmarks has become critical. However, existing benchmarks focus more on code generation tasks in function and class level and provide rich text description to prompt the model. By contrast, such descriptive prompt is commonly unavailable in real development and code completion can occur in wider range of situations such as in the middle of a function or a code block. These limitations makes the evaluation poorly align with the practical scenarios of code completion tools. In this paper, we propose RepoMasterEval, a novel benchmark for evaluating code completion models constructed from real-world Python and TypeScript repositories. Each benchmark datum is generated by masking a code snippet (ground truth) from one source code file with existing test suites. To improve test accuracy of model generated code, we employ mutation testing to measure the effectiveness of the test cases and we manually crafted new test cases for those test suites with low mutation score. Our empirical evaluation on 6 state-of-the-art models shows that test argumentation is critical in improving the accuracy of the benchmark and RepoMasterEval is able to report difference in model performance in real-world scenarios. The deployment of RepoMasterEval in a collaborated company for one month also revealed that the benchmark is useful to give accurate feedback during model training and the score is in high correlation with the model's performance in practice. Based on our findings, we call for the software engineering community to build more LLM benchmarks tailored for code generation tools taking the practical and complex development environment into consideration.

Many tasks revolve around editing a document, whether code or text. We formulate the revision similarity problem to unify a wide range of machine learning evaluation problems whose goal is to assess a revision to an existing document. We observe that revisions usually change only a small portion of an existing document, so the existing document and its immediate revisions share a majority of their content. We formulate five adequacy criteria for revision similarity measures, designed to align them with human judgement. We show that popular pairwise measures, like BLEU, fail to meet these criteria, because their scores are dominated by the shared content. They report high similarity between two revisions when humans would assess them as quite different. This is a fundamental flaw we address. We propose a novel static measure, Excision Score (ES), which computes longest common subsequence (LCS) to remove content shared by an existing document with the ground truth and predicted revisions, before comparing only the remaining divergent regions. This is analogous to a surgeon creating a sterile field to focus on the work area. We use approximation to speed the standard cubic LCS computation to quadratic. In code-editing evaluation, where static measures are often used as a cheap proxy for passing tests, we demonstrate that ES surpasses existing measures. When aligned with test execution on HumanEvalFix, ES improves over its nearest competitor, SARI, by 12% Pearson correlation and by >21% over standard measures like BLEU. The key criterion is invariance to shared context; when we perturb HumanEvalFix with increased shared context, ES' improvement over SARI increases to 20% and >30% over standard measures. ES also handles other corner cases that other measures do not, such as correctly aligning moved code blocks, and appropriately rewarding matching insertions or deletions.

Zero-shot LLMs are now also used for textual classification tasks, e.g., sentiment/emotion detection of a given input as a sentence/article. However, their performance can be suboptimal in such data annotation tasks. We introduce a novel technique Perceived Confidence Scoring (PCS) that evaluates LLM's confidence for its classification of an input by leveraging Metamorphic Relations (MRs). The MRs generate semantically equivalent yet textually mutated versions of the input. Following the principles of Metamorphic Testing (MT), the mutated versions are expected to have annotation labels similar to the input. By analyzing the consistency of LLM responses across these variations, PCS computes a confidence score based on the frequency of predicted labels. PCS can be used both for single LLM and multiple LLM settings (e.g., majority voting). We introduce an algorithm Perceived Differential Evolution (PDE) that determines the optimal weights assigned to the MRs and the LLMs for a classification task. Empirical evaluation shows PCS significantly improves zero-shot accuracy for Llama-3-8B-Instruct (4.96%) and Mistral-7B-Instruct-v0.3 (10.52%), with Gemma-2-9b-it showing a 9.39% gain. When combining all three models, PCS significantly outperforms majority voting by 7.75%.

Protein mutation effect generation asks a model to describe the functional consequence of a point mutation in natural language. Existing protein-to-text systems typically encode mutation information into undifferentiated representations, overlooking the organization of mutation-induced evidence across structural and biochemical factors. We propose RipplePLM, a mutation-aware generation framework centered on Direct-Distal Cross-Attention (DDCA). By constructing a residue-level Mutation Perturbation Field from pre-trained protein language models, DDCA leverages predicted contact maps to organize mutation representations into two pathways: the mutation site's immediate contact neighborhood and its multi-hop distal context. To complement this structural decomposition, we further introduce the Property Latent Chain (PLChain), which injects expert-guided supervision of biochemical property changes (e.g., thermostability and optimal pH) into the LLM hidden-state pathway through latent property tokens. On MutaDescribe, RipplePLM improves over mutation-specific baselines on temporal and structural splits; under a matched-backbone comparison, average structural-split ROUGE-L increases from 22.23 to 35.65. Expert evaluation further shows a higher proportion of biologically accurate or relevant descriptions than the mutation-specific baseline. Additional ablations, representation diagnostics, and low-N fitness regression experiments further support the effectiveness of the learned mutation-aware representations. Code: https://github.com/Lyu6PosHao/RipplePLM.

Protein Language Models (PLMs) have emerged as performant and scalable tools for predicting the functional impact and clinical significance of protein-coding variants, but they still lag experimental accuracy. Here, we present a novel fine-tuning approach to improve the performance of PLMs with experimental maps of variant effects from Deep Mutational Scanning (DMS) assays using a Normalised Log-odds Ratio (NLR) head. We find consistent improvements in a held-out protein test set, and on independent DMS and clinical variant annotation benchmarks from ProteinGym and ClinVar. These findings demonstrate that DMS is a promising source of sequence diversity and supervised training data for improving the performance of PLMs for variant effect prediction.

This paper describes Meta's ACH system for mutation-guided LLM-based test generation. ACH generates relatively few mutants (aka simulated faults), compared to traditional mutation testing. Instead, it focuses on generating currently undetected faults that are specific to an issue of concern. From these currently uncaught faults, ACH generates tests that can catch them, thereby `killing' the mutants and consequently hardening the platform against regressions. We use privacy concerns to illustrate our approach, but ACH can harden code against {\em any} type of regression. In total, ACH was applied to 10,795 Android Kotlin classes in 7 software platforms deployed by Meta, from which it generated 9,095 mutants and 571 privacy-hardening test cases. ACH also deploys an LLM-based equivalent mutant detection agent that achieves a precision of 0.79 and a recall of 0.47 (rising to 0.95 and 0.96 with simple pre-processing). ACH was used by Messenger and WhatsApp test-a-thons where engineers accepted 73% of its tests, judging 36% to privacy relevant. We conclude that ACH hardens code against specific concerns and that, even when its tests do not directly tackle the specific concern, engineers find them useful for their other benefits.

Despite being self-supervised, protein language models have shown remarkable performance in fundamental biological tasks such as predicting impact of genetic variation on protein structure and function. The effectiveness of these models on diverse set of tasks suggests that they learn meaningful representations of fitness landscape that can be useful for downstream clinical applications. Here, we interrogate the use of these language models in characterizing known pathogenic mutations in curated, medically actionable genes through an exhaustive search of putative compensatory mutations on each variant's genetic background. Systematic analysis of the predicted effects of these compensatory mutations reveal unappreciated structural features of proteins that are missed by other structure predictors like AlphaFold. While deep mutational scan experiments provide an unbiased estimate of the mutational landscape, we encourage the community to generate and curate rescue mutation experiments to inform the design of more sophisticated co-masking strategies and leverage large language models more effectively for downstream clinical prediction tasks.

Efficient bug resolution is critical for maintaining software quality and user satisfaction. However, specific bug reports experience unusually long resolution times, which may indicate underlying process inefficiencies or complex issues. This study presents a comprehensive analysis of bug resolution anomalies across seven prominent open-source repositories: Cassandra, Firefox, Hadoop, HBase, SeaMonkey, Spark, and Thunderbird. Utilizing statistical methods such as Z-score and Interquartile Range (IQR), we identify anomalies in bug resolution durations. To understand the thematic nature of these anomalies, we apply Term Frequency-Inverse Document Frequency (TF-IDF) for textual feature extraction and KMeans clustering to group similar bug summaries. Our findings reveal consistent patterns across projects, with anomalies often clustering around test failures, enhancement requests, and user interface issues. This approach provides actionable insights for project maintainers to prioritize and effectively address long-standing bugs.

Predicting the fitness impact of mutations is central to protein engineering but constrained by limited assays relative to the size of sequence space. Protein language models (pLMs) trained with masked language modeling (MLM) exhibit strong zero-shot fitness prediction; we provide a unifying view by interpreting natural evolution as implicit reward maximization and MLM as inverse reinforcement learning (IRL), in which extant sequences act as expert demonstrations and pLM log-odds serve as fitness estimates. Building on this perspective, we introduce EvoIF, a lightweight model that integrates two complementary sources of evolutionary signal: (i) within-family profiles from retrieved homologs and (ii) cross-family structural-evolutionary constraints distilled from inverse folding logits. EvoIF fuses sequence-structure representations with these profiles via a compact transition block, yielding calibrated probabilities for log-odds scoring. On ProteinGym (217 mutational assays; >2.5M mutants), EvoIF and its MSA-enabled variant achieve state-of-the-art or competitive performance while using only 0.15% of the training data and fewer parameters than recent large models. Ablations confirm that within-family and cross-family profiles are complementary, improving robustness across function types, MSA depths, taxa, and mutation depths. The codes will be made publicly available.

Language models (LMs) perform well on standardized coding benchmarks but struggle with real-world software engineering tasks such as resolving GitHub issues in SWE-Bench, especially when model parameters are less than 100B. While smaller models are preferable in practice due to their lower computational cost, improving their performance remains challenging. Existing approaches primarily rely on supervised fine-tuning (SFT) with high-quality data, which is expensive to curate at scale. An alternative is test-time scaling: generating multiple outputs, scoring them using a verifier, and selecting the best one. Although effective, this strategy often requires excessive sampling and costly scoring, limiting its practical application. We propose Evolutionary Test-Time Scaling (EvoScale), a sample-efficient method that treats generation as an evolutionary process. By iteratively refining outputs via selection and mutation, EvoScale shifts the output distribution toward higher-scoring regions, reducing the number of samples needed to find correct solutions. To reduce the overhead from repeatedly sampling and selection, we train the model to self-evolve using reinforcement learning (RL). Rather than relying on external verifiers at inference time, the model learns to self-improve the scores of its own generations across iterations. Evaluated on SWE-Bench-Verified, EvoScale enables our 32B model, Satori-SWE-32B, to match or exceed the performance of models with over 100B parameters while using a few samples. Code, data, and models will be fully open-sourced.

Measuring diversity accurately is important for many scientific fields, including machine learning (ML), ecology, and chemistry. The Vendi Score was introduced as a generic similarity-based diversity metric that extends the Hill number of order q=1 by leveraging ideas from quantum statistical mechanics. Contrary to many diversity metrics in ecology, the Vendi Score accounts for similarity and does not require knowledge of the prevalence of the categories in the collection to be evaluated for diversity. However, the Vendi Score treats each item in a given collection with a level of sensitivity proportional to the item's prevalence. This is undesirable in settings where there is a significant imbalance in item prevalence. In this paper, we extend the other Hill numbers using similarity to provide flexibility in allocating sensitivity to rare or common items. This leads to a family of diversity metrics -- Vendi scores with different levels of sensitivity -- that can be used in a variety of applications. We study the properties of the scores in a synthetic controlled setting where the ground truth diversity is known. We then test their utility in improving molecular simulations via Vendi Sampling. Finally, we use the Vendi scores to better understand the behavior of image generative models in terms of memorization, duplication, diversity, and sample quality.

Genomic language models (gLMs) hold promise for generating novel, functional DNA sequences for synthetic biology. However, realizing this potential requires models to go beyond evolutionary plausibility and understand how DNA sequence encodes gene expression and regulation. We introduce a benchmark called Nullsettes, which assesses how well models can predict in silico loss-of-function (LOF) mutations, in synthetic expression cassettes with little evolutionary precedent. Testing 12 state-of-the-art gLMs, we find that most fail to consistently detect these strong LOF mutations. All models show a sharp drop in predictive accuracy as the likelihood assigned to the original (nonmutant) sequence decreases, suggesting that gLMs rely heavily on pattern-matching to their evolutionary prior rather than on any mechanistic understanding of gene expression. Our findings highlight fundamental limitations in how gLMs generalize to engineered, non-natural sequences, and underscore the need for benchmarks and modeling strategies that prioritize functional understanding.

Neurological diseases are the leading global cause of disability, yet most lack disease-modifying treatments. We present PROTON, a heterogeneous graph transformer that generates testable hypotheses across molecular, organoid, and clinical systems. To evaluate PROTON, we apply it to Parkinson's disease (PD), bipolar disorder (BD), and Alzheimer's disease (AD). In PD, PROTON linked genetic risk loci to genes essential for dopaminergic neuron survival and predicted pesticides toxic to patient-derived neurons, including the insecticide endosulfan, which ranked within the top 1.29% of predictions. In silico screens performed by PROTON reproduced six genome-wide α-synuclein experiments, including a split-ubiquitin yeast two-hybrid system (normalized enrichment score [NES] = 2.30, FDR-adjusted p < 1 times 10^{-4}), an ascorbate peroxidase proximity labeling assay (NES = 2.16, FDR < 1 times 10^{-4}), and a high-depth targeted exome sequencing study in 496 synucleinopathy patients (NES = 2.13, FDR < 1 times 10^{-4}). In BD, PROTON predicted calcitriol as a candidate drug that reversed proteomic alterations observed in cortical organoids derived from BD patients. In AD, we evaluated PROTON predictions in health records from n = 610,524 patients at Mass General Brigham, confirming that five PROTON-predicted drugs were associated with reduced seven-year dementia risk (minimum hazard ratio = 0.63, 95% CI: 0.53-0.75, p < 1 times 10^{-7}). PROTON generated neurological hypotheses that were evaluated across molecular, organoid, and clinical systems, defining a path for AI-driven discovery in neurological disease.

Deep learning has been widely used for protein engineering. However, it is limited by the lack of sufficient experimental data to train an accurate model for predicting the functional fitness of high-order mutants. Here, we develop SESNet, a supervised deep-learning model to predict the fitness for protein mutants by leveraging both sequence and structure information, and exploiting attention mechanism. Our model integrates local evolutionary context from homologous sequences, the global evolutionary context encoding rich semantic from the universal protein sequence space and the structure information accounting for the microenvironment around each residue in a protein. We show that SESNet outperforms state-of-the-art models for predicting the sequence-function relationship on 26 deep mutational scanning datasets. More importantly, we propose a data augmentation strategy by leveraging the data from unsupervised models to pre-train our model. After that, our model can achieve strikingly high accuracy in prediction of the fitness of protein mutants, especially for the higher order variants (> 4 mutation sites), when finetuned by using only a small number of experimental mutation data (<50). The strategy proposed is of great practical value as the required experimental effort, i.e., producing a few tens of experimental mutation data on a given protein, is generally affordable by an ordinary biochemical group and can be applied on almost any protein.

Time series agents answer analytical questions by calling external tools, and which tools they carry is decided by people before the agent runs. However, we identify two failures in this setup. Human-Agent Tool Misalignment: a library of 21 expert-curated tools helps on some tasks and hurts on others, dropping anomaly accuracy under every backbone we test. Silent Harm: one round of generic self-revision changes 147 answers and breaks 56 of them, while the final score moves by less than a point. Both follow from the same gap: whether a tool helps is decided question by question at runtime, while tools are supplied in advance and judged by a single average. To address this, we propose TimeEvo, which clusters an agent's diagnosed failures into capability gaps, plans a measurement for each, synthesizes evidence-only tools that fill them, and admits the candidate library only through a paired admission gate. Experiments on ten time series QA tasks and three backbones show that TimeEvo, starting from an empty library, improves accuracy on every task and every backbone, and that a library grown on a cheap model still gains when it is installed into stronger ones. Code is available at https://github.com/Muyiiiii/TimeEvo.

Large language models can now write complete, interactive 3D worlds as code, but grading those worlds automatically is unreliable. Existing judges take one view of the output: a vision-language model scores a few rendered snapshots, or a language model reads the source. On worlds written by five frontier models we find that the two views disagree on 32% of required items, mostly code that no frame shows, and that fixed views miss small close-up contents. We present WorldBench, a benchmark and judge for open-ended, LLM-generated Three.js worlds. From one prompt describing a floating voxel island with ten biomes, physics, and day/night and seasonal cycles, the judge explores the running world, controlling its clock, orbiting it, and sending a navigator agent to frame each biome, and reads the code for what it sees. Neither channel is trusted on its own: a code quote counts only if it is text the source contains, and visual claims are checked against measured pixels where the property is measurable. A mutation test, in which we remove features by construction, shows that code-only judging gives full credit to four of five removed features, because their code remains in the file. Our judge cuts the points kept on removed features by a third (5.44 to 3.55 of 7.11), and what it still credits is mostly code that exists but never runs. We evaluate five frontier models: Claude Fable 5.1, GPT-6 Astra, Kimi K3, Grok 4.7 and Gemini 3.1 Pro. Code, prompt, tests and judge configuration are available at https://github.com/KrishBakshi/worldbench

Nucleotide sequence variation can induce significant shifts in functional fitness. Recent nucleotide foundation models promise to predict such fitness effects directly from sequence, yet heterogeneous datasets and inconsistent preprocessing make it difficult to compare methods fairly across DNA and RNA families. Here we introduce NABench, a large-scale, systematic benchmark for nucleic acid fitness prediction. NABench aggregates 162 high-throughput assays and curates 2.6 million mutated sequences spanning diverse DNA and RNA families, with standardized splits and rich metadata. We show that NABench surpasses prior nucleotide fitness benchmarks in scale, diversity, and data quality. Under a unified evaluation suite, we rigorously assess 29 representative foundation models across zero-shot, few-shot prediction, transfer learning, and supervised settings. The results quantify performance heterogeneity across tasks and nucleic-acid types, demonstrating clear strengths and failure modes for different modeling choices and establishing strong, reproducible baselines. We release NABench to advance nucleic acid modeling, supporting downstream applications in RNA/DNA design, synthetic biology, and biochemistry. Our code is available at https://github.com/mrzzmrzz/NABench.

Scientific agents increasingly analyze data, execute code, and produce research artifacts, yet most benchmarks emphasize final answers, isolated programs, or a single domain. We introduce FrontierChallenge, a cross-domain benchmark comprising 300 end-to-end scientific workflows. In this paper, we release and evaluate 97 of these tasks, spanning quantum chemistry, molecular dynamics, materials characterization, analytical chemistry, life science, and electrochemistry/environment. Each task provides fixed inputs and specifies a bundle of required scientific deliverables. We evaluate twelve frontier models with three agent scaffolds. Pass Rate measures the fraction of tasks satisfying the full-completion criterion, while Avg. Score captures partial progress. Each of the best-performing configurations completed only 20 of the 97 released tasks, yielding a Pass Rate of 20.6%. Partial progress translated especially poorly into complete delivery in analytical chemistry and electrochemistry/environment: Avg. Scores reached 87.6 and 94.9, but the highest Pass Rates were only 4% and 0%. Among non-passing Claude Code trajectories, 75.5% still ended with language claiming completion. These findings show that neither high partial scores nor confident claims of completion reliably indicate that a scientific task has been fully delivered, highlighting the need to evaluate end-to-end workflow execution and the completeness of scientific deliverables together.

Repository-level performance-optimization benchmarks such as GSO, SWE-Perf and SWE-fficiency evaluate coding agents by applying patches to real repositories and comparing runtime against unoptimized baselines and official reference patches. Their leaderboard scores are increasingly used as evidence of coding-agent progress, but those scores can conflate runtime instability, benchmark-specific scoring rules, and how many tasks are already solved by at least one public submission. We audit these issues across the three benchmarks. First, we replay the official reference patches for 740 code optimization tasks across four common types of Google Cloud machines. Most benchmark tasks can be replayed, but their reference patches satisfy the original benchmark validity rules in every cross-machine replay for only 39/102 GSO tasks, 11/140 SWE-Perf tasks, and 411/498 SWE-fficiency tasks; SWE-Perf is especially fragile because many reference patches produce close-to-zero runtime changes. Second, we show that public submission rankings depend strongly on the benchmark scoring rule. Among eight public submissions shared by GSO and SWE-fficiency, the official rankings disagree on 9 of 28 pairwise submission comparisons, and SWE-fficiency's leaderboard scoring rule assigns the worst ten tasks overly high score weights of 58.5%-82.8%. Third, looking across 10 public submissions for each task, we find that at least one submission matches or beats the reference patch on 85.3% (384/450) of replay-valid GSO and SWE-fficiency tasks, and beats the unoptimized base code on 99.8% (449/450). Our study complements leaderboard scores by identifying tasks with more reliable performance signals, quantifying per-task score contributions, and exposing the remaining performance gaps that are hidden by aggregate rankings.

Evolutionary search with large language models (LLMs) can stall when progress requires external knowledge the model lacks. Supplying relevant documents helps, but simply adding web search tool can keep returning the same pages as solutions change. We introduce EvoDuet, a bi-level optimization method that co-evolves solutions and search queries with fixed model parameters. At each iteration, a retrieval gate lets the LLM assess its knowledge gap and choose to retrieve new documents, reuse stored ones, or proceed without them. An inner loop refines queries and ranks documents by the solution scores they are predicted to yield; an outer loop generates candidates in parallel from these documents and records the evaluated outcomes for later searches. Across 21 optimization tasks with one candidate per iteration, EvoDuet raises OpenEvolve's normalized discovery gain from 74.1% to 78.0% with GPT-5.6-Luna and from 61.3% to 82.3% with Gemini-3.8-Flash, whereas Qwen3.5-9B does not benefit. Our best runs surpass the previously reported best scores on eight tasks, including Swap Reduction on Q20 and Rosetta, and match them on three more. EvoDuet also improves with other scaffolds (e.g., Top-K, EvoX) on Sums/Diffs and Denoising, demonstrating its applicability across evolutionary search scaffolds.

RTL design often relies heavily on ad-hoc testbench creation early in the design cycle. While large language models (LLMs) show promise for RTL code generation, their ability to reason about hardware specifications and generate targeted test plans remains largely unexplored. We present the first systematic study of LLM reasoning capabilities for RTL verification stimuli generation, establishing a two-stage framework that decomposes test plan generation from testbench execution. Our benchmark reveals that state-of-the-art models, including DeepSeek-R1 and Claude-4.0-Sonnet, achieve only 15.7-21.7% success rates on generating stimuli that pass golden RTL designs. To improve LLM generated stimuli, we develop a comprehensive training methodology combining supervised fine-tuning with a novel reinforcement learning approach, GRPO with State Mutation (GRPO-SMu), which enhances exploration by varying input mutations. Our approach leverages a tree-based branching mutation strategy to construct training data comprising equivalent and mutated trees, moving beyond linear mutation approaches to provide rich learning signals. Training on this curated dataset, our 7B parameter model achieves a 33.3% golden test pass rate and a 13.9% mutation detection rate, representing a 17.6% absolute improvement over baseline and outperforming much larger general-purpose models. These results demonstrate that specialized training methodologies can significantly enhance LLM reasoning capabilities for hardware verification tasks, establishing a foundation for automated sub-unit testing in semiconductor design workflows.

Stabilizing proteins is a foundational step in protein engineering. However, the evolutionary pressure of all extant proteins makes identifying the scarce number of mutations that will improve thermodynamic stability challenging. Deep learning has recently emerged as a powerful tool for identifying promising mutations. Existing approaches, however, are computationally expensive, as the number of model inferences scales with the number of mutations queried. Our main contribution is a simple, parallel decoding algorithm. Our Mutate Everything is capable of predicting the effect of all single and double mutations in one forward pass. It is even versatile enough to predict higher-order mutations with minimal computational overhead. We build Mutate Everything on top of ESM2 and AlphaFold, neither of which were trained to predict thermodynamic stability. We trained on the Mega-Scale cDNA proteolysis dataset and achieved state-of-the-art performance on single and higher-order mutations on S669, ProTherm, and ProteinGym datasets. Code is available at https://github.com/jozhang97/MutateEverything

Agent benchmarks often evaluate only final answers even when agents run on stateful runtimes. We argue this under-specifies what is being evaluated: the proper unit is a declared model-plus-runtime configuration whose failures can occur in evidence acquisition, runtime routing, safety boundaries, or repeated execution. We present ClawProBench, a trace-aware benchmark for runtime-native agent evaluation instantiated on OpenClaw, a live agent runtime with workspace tools and native surfaces for browsing, memory, messaging, scheduling, skills, and subagents. ClawProBench defines two tracks: a 102-scenario full profile with live workspace and native-runtime routing tasks, and a frozen 68-scenario holdout with closed-world JSON output contracts for robust ranking. Trials are scored from execution traces via a safety-gated formula combining correctness, process quality, and efficiency, preserving failure evidence for audit. Our anonymous artifact includes benchmark definitions, scoring code, manifests and sanitized traces. We evaluate 68 configurations on the full profile and 37 on holdout. The top safety-gated average trace score is 0.7671. Native-runtime tasks underperform workspace-live tasks (0.5238 vs. 0.6415). On holdout, pass@k-any outperforms strict three-trial pass (0.6638 vs. 0.2890), while full-profile and holdout rankings show weak alignment (Spearman 0.1300). Rankings based purely on correctness differ substantially from process-aware, safety-gated and strict-pass views. Final-answer leaderboards may hide native-surface weaknesses, one-off successes and trace-local agent failure modes.

Routine complete blood counts (CBCs) could yield new biomarkers, but the private records needed to evaluate candidates cannot be shared with frontier language model agents that excel at discovery. We distilled the evidence held in the Clalit Health Services panel of over 5.4 million patients into a released scoring tool: for each of 13 immune-mediated diseases, a graph attention network was trained inside the data boundary to predict the case-control AUC of candidate CBC expressions, and only the trained weights were released. The tool grounds an agent's propose-score-refine loop in real-world data without exposing any patient data. In external validation, agent-discovered expressions improved on their literature-seeded starting points by a median of 4.18 AUC percentage points, and across three independent cohorts, reranking the candidates of three frontier research tools improved on their first choices in most comparisons, with gains that varied by cohort. The released scorer supports privacy-preserving biomarker hypothesis generation.

Large language models (LLMs) have shown impressive promise in code generation, yet their progress remains limited by the shortage of large-scale datasets that are both diverse and well-aligned with human reasoning. Most existing resources pair problems with solutions, but omit the intermediate thought process that guides coding. To close this gap, we present a scalable synthetic data generation pipeline that produces nearly 800k instruction-reasoning-code-test quadruplets. Each sample combines a task, a step-by-step reasoning trace, a working solution, and executable tests, enabling models to learn not just the what but also the how of problem solving. Our pipeline combines four key components: curated contest problems, web-mined content filtered by relevance classifiers, data expansion guided by reasoning patterns, and multi-stage execution-based validation. A genetic mutation algorithm further increases task diversity while maintaining consistency between reasoning traces and code implementations. Our key finding is that fine-tuning LLMs on this dataset yields consistent improvements on coding benchmarks. Beyond raw accuracy, reasoning-aware data can substitute for model scaling, generalize across architectures, and outperform leading open-source alternatives under identical sample budgets. Our work establishes reasoning-centered synthetic data generation as an efficient approach for advancing coding capabilities in LLMs. We publish our dataset and generation pipeline to facilitate further research.

During a virus's evolution,various regions of the genome are subjected to distinct levels of functional constraints.Combined with factors like codon bias and DNA repair efficiency,these constraints contribute to unique mutation patterns within the genome or a specific gene. In this project, we harnessed the power of Large Language Models(LLMs) to predict the evolution of SARS-CoV-2. By treating the mutation process from one generation to the next as a translation task, we trained a transformer model, called VirusT5, to capture the mutation patterns underlying SARS-CoV-2 evolution. We evaluated the VirusT5's ability to detect these mutation patterns including its ability to identify mutation hotspots and explored the potential of using VirusT5 to predict future virus variants. Our findings demonstrate the feasibility of using a large language model to model viral evolution as a translation process. This study establishes the groundbreaking concept of "mutation-as-translation," paving the way for new methodologies and tools for combating virus threats

Scientific ideas rarely start from a blank page. They inherit mechanisms, repair known limitations, and recombine pieces of earlier work, much like biological genomes. Current benchmarks still say little about whether AI systems can follow this inheritance structure. We present IdeaGene-Bench (IG-Bench), a benchmark for scientific lineage reasoning and lineage-grounded idea generation. IG-Bench is organized around the IdeaGene framework: each paper or proposal is represented as a set of minimal, typed, evidence-grounded Idea Genome objects, and a GenomeDiff aligns these objects to record inheritance, mutation, loss, external import, and novel insertion under six operational evolutionary dynamics. The benchmark contains 1,961 golden lineage traces, 1,085 curated Idea Genome objects, and 920 pairwise GenomeDiff records across 10 scientific domains. It supports two evaluations. IG-Exam (42 task types, 1,029 instances) tests closed-form lineage reasoning across Idea Genome abstraction, inheritance tracing, evolutionary reasoning, and lineage verification. IG-Arena evaluates generation with a lineage-conditioned Population-Evolution Score(PES), asking whether a proposal can be inserted as a coherent descendant of a given lineage population: it should inherit the right Idea Genome objects, vary meaningfully from nearby work, and offer selection value for future research. Experiments on 14 LLM-based scientists expose a compositional bottleneck. The strongest system reaches only 27.3% exact accuracy on lineage reasoning, and structured lineage context reshuffles system rankings rather than helping every participant uniformly.

We study the problem of optimizing biological sequences, e.g., proteins, DNA, and RNA, to maximize a black-box score function that is only evaluated in an offline dataset. We propose a novel solution, bootstrapped training of score-conditioned generator (BootGen) algorithm. Our algorithm repeats a two-stage process. In the first stage, our algorithm trains the biological sequence generator with rank-based weights to enhance the accuracy of sequence generation based on high scores. The subsequent stage involves bootstrapping, which augments the training dataset with self-generated data labeled by a proxy score function. Our key idea is to align the score-based generation with a proxy score function, which distills the knowledge of the proxy score function to the generator. After training, we aggregate samples from multiple bootstrapped generators and proxies to produce a diverse design. Extensive experiments show that our method outperforms competitive baselines on biological sequential design tasks. We provide reproducible source code: https://github.com/kaist-silab/bootgen{https://github.com/kaist-silab/bootgen}.

DNA-synthesis providers screen incoming orders by searching the requested sequence against curated hazard lists. We show that this baseline collapses to a 100% false-flag rate when the hazardous sequence comes from a taxonomic family absent from the reference set: under Conformal Risk Control's certified miss-rate constraint, a low-discrimination signal forces the threshold below the entire test-benign mass. We compose three signals derived from a synthesis order's public annotation: k-mer Jaccard similarity to known toxins, the trimmed-mean score of a five-LLM judge panel, and cosine similarity to clustered embedding centroids. Fused under a monotone logistic aggregator and calibrated by Conformal Risk Control, the resulting screener certifies E[FNR] le α. Across ten leave-one-taxonomic-family-out folds at α=0.05 on UniProt KW-0800 reviewed toxins, the calibrated screener achieves 0% test miss rate on every fold and 0% test false-flag rate on nine of ten folds. The bound's finite-sample slack 1/(n_{cal}+1) caps the certifiable miss rate at 1.77% on our 200-hazard subsample; reaching procurement-grade α=10^{-3} requires an 18times larger calibration set, which the full reviewed UniProt KW-0800 corpus is large enough to deliver. The binding constraint on certifiable DNA-synthesis screening is calibration data, not algorithms. Code: https://github.com/najmulhasan-code/crc-screen

We present RM-RF, a lightweight reward model for run-free evaluation of automatically generated unit tests. Instead of repeatedly compiling and executing candidate tests, RM-RF predicts - from source and test code alone - three execution-derived signals: (1) whether the augmented test suite compiles and runs successfully, (2) whether the generated test cases increase code coverage, and (3) whether the generated test cases improve the mutation kill rate. To train and evaluate RM-RF we assemble a multilingual dataset (Java, Python, Go) of focal files, test files, and candidate test additions labeled by an execution-based pipeline, and we release an associated dataset and methodology for comparative evaluation. We tested multiple model families and tuning regimes (zero-shot, full fine-tuning, and PEFT via LoRA), achieving an average F1 of 0.69 across the three targets. Compared to conventional compile-and-run instruments, RM-RF provides substantially lower latency and infrastructure cost while delivering competitive predictive fidelity, enabling fast, scalable feedback for large-scale test generation and RL-based code optimization.

This beta technical report asks how reusable experience should be represented so that it can function as effective test-time control and as a substrate for iterative evolution. We study this question in 4.590 controlled trials across 45 scientific code-solving scenarios. We find that documentation-oriented Skill packages provide unstable control: their useful signal is sparse, and expanding a compact experience object into a fuller documentation package often fails to help and can degrade the overall average. We further show that representation itself is a first-order factor. A compact Gene representation yields the strongest overall average, remains competitive under substantial structural perturbations, and outperforms matched-budget Skill fragments, while reattaching documentation-oriented material usually weakens rather than improves it. Beyond one-shot control, we show that Gene is also a better carrier for iterative experience accumulation: attached failure history is more effective in Gene than in Skill or freeform text, editable structure matters beyond content alone, and failure information is most useful when distilled into compact warnings rather than naively appended. On CritPt, gene-evolved systems improve over their paired base models from 9.1% to 18.57% and from 17.7% to 27.14%. These results suggest that the core problem in experience reuse is not how to supply more experience, but how to encode experience as a compact, control-oriented, evolution-ready object.

Background: Cancer remains one of the leading causes of morbidity and mortality worldwide. Comprehensive datasets that combine histopathological images with genetic and survival data across various tumour sites are essential for advancing computational pathology and personalised medicine. Results: We present SurGen, a dataset comprising 1,020 H&E-stained whole slide images (WSIs) from 843 colorectal cancer cases. The dataset includes detailed annotations for key genetic mutations (KRAS, NRAS, BRAF) and mismatch repair status, as well as survival data for 426 cases. To demonstrate SurGen's practical utility, we conducted a proof-of-concept machine learning experiment predicting mismatch repair status from the WSIs, achieving a test AUROC of 0.8316. These preliminary results underscore the dataset's potential to facilitate research in biomarker discovery, prognostic modelling, and advanced machine learning applications in colorectal cancer. Conclusions: SurGen offers a valuable resource for the scientific community, enabling studies that require high-quality WSIs linked with comprehensive clinical and genetic information on colorectal cancer. Our initial findings affirm the dataset's capacity to advance diagnostic precision and foster the development of personalised treatment strategies in colorectal oncology. Data available online at https://doi.org/10.6019/S-BIAD1285.

Visual artifacts remain a persistent challenge in diffusion models, even with training on massive datasets. Current solutions primarily rely on supervised detectors, yet lack understanding of why these artifacts occur in the first place. In our analysis, we identify three distinct phases in the diffusion generative process: Profiling, Mutation, and Refinement. Artifacts typically emerge during the Mutation phase, where certain regions exhibit anomalous score dynamics over time, causing abrupt disruptions in the normal evolution pattern. This temporal nature explains why existing methods focusing only on spatial uncertainty of the final output fail at effective artifact localization. Based on these insights, we propose ASCED (Abnormal Score Correction for Enhancing Diffusion), that detects artifacts by monitoring abnormal score dynamics during the diffusion process, with a trajectory-aware on-the-fly mitigation strategy that appropriate generation of noise in the detected areas. Unlike most existing methods that apply post hoc corrections, \eg, by applying a noising-denoising scheme after generation, our mitigation strategy operates seamlessly within the existing diffusion process. Extensive experiments demonstrate that our proposed approach effectively reduces artifacts across diverse domains, matching or surpassing existing supervised methods without additional training.

Probes are the workhorse of interpretability. If a model's hidden states predict a variable, the model is said to represent it. But a probe score has no fixed meaning. An R^2 of 0.6 may only reflect what the input already gives away, and the same score can mean different things on different data. We propose reading every probe score against two reference points: a floor, what a declared set of simple inputs already predicts, and a ceiling, what the full input can predict. The gap between them, the headroom, is the range in which a probe can show that a model computes something beyond the simple inputs. We prove that headroom vanishes in two ways: the target stops depending on a hidden variable the model must infer, or the input stops revealing it. We test this on transformers trained for in-context meta-analysis, which must infer the hidden heterogeneity between studies to weight them correctly, and where both reference points are known. Under distribution shift, probe scores fall and prediction error rises 12--15times, yet the model recovers a similar share of the headroom, indicating that the data lost information, not the representation. We then analyze the real models. The single-cell foundation model scGPT encodes biological variability only partially. We also revisit four influential LLM probing studies, which claim that models represent geography, the state of an Othello board, truth, and the demographics of their users. Against a floor computed from the input text alone, some of these claims hold, while others are largely explained by the text itself.

In passage reranking, response ranking and multi-document question answering, LLMs can score several candidate documents or responses together in one prompt, each still receiving its own score. Such scorers are selected on ranking quality, but their scores determine a decision: what a score threshold retains, a reader answers, or which chosen/rejected pair enters preference training. Because the candidates share that prompt, reordering them changes their scores. The same query over the same candidates should still yield the same decision. However, equal ranking quality does not imply equal decisions: on passage reranking, five trained scorers within 0.010 nDCG@10 retain sets that overlap by only 0.66-0.84 when reordered. No prompt-time change we test resolves that dependence: the only one that improves ranking quality does not measurably improve decision stability. We introduce order-consistency SFT (OC-SFT), which attenuates it in the weights by penalizing disagreement between a candidate's scores across orderings. It holds ranking quality and leads every decision-stability measure among trained scorers on all three tasks. It is also more stable on 12 base models than order-averaged distillation, which trains on labels averaged across permutations. One OC-SFT permutation retains sets that overlap more than ten averaged off-the-shelf permutations. A comparison of such scorers should therefore report what a threshold retains and a reader answers, not ranking quality alone. Code is available at https://github.com/thomsonreuters/presentation-dependence.

Large language models (LLMs) have achieved expert-level performance on standardized examinations, yet multiple-choice accuracy poorly reflects real-world clinical utility and safety. As patients and clinicians increasingly use LLMs for guidance on complex conditions such as pancreatic cancer, evaluation must extend beyond general medical knowledge. Existing frameworks, such as HealthBench, rely on simulated queries and lack disease-specific depth. Moreover, high rubric-based scores do not ensure factual correctness, underscoring the need to assess hallucinations. We developed a human-in-the-loop pipeline to create expert rubrics for de-identified patient questions from the Pancreatic Cancer Action Network (PanCAN). The resulting benchmark, PanCanBench, includes 3,130 question-specific criteria across 282 authentic patient questions. We evaluated 22 proprietary and open-source LLMs using an LLM-as-a-judge framework, measuring clinical completeness, factual accuracy, and web-search integration. Models showed substantial variation in rubric-based completeness, with scores ranging from 46.5% to 82.3%. Factual errors were common, with hallucination rates (the percentages of responses containing at least one factual error) ranging from 6.0% for Gemini-2.5 Pro and GPT-4o to 53.8% for Llama-3.1-8B. Importantly, newer reasoning-optimized models did not consistently improve factuality: although o3 achieved the highest rubric score, it produced inaccuracies more frequently than other GPT-family models. Web-search integration did not inherently guarantee better responses. The average score changed from 66.8% to 63.9% for Gemini-2.5 Pro and from 73.8% to 72.8% for GPT-5 when web search was enabled. Synthetic AI-generated rubrics inflated absolute scores by 17.9 points on average while generally maintaining similar relative ranking.

In recent years, the use of automated source code generation utilizing transformer-based generative models has expanded, and these models can generate functional code according to the requirements of the developers. However, recent research revealed that these automatically generated source codes can contain vulnerabilities and other quality issues. Despite researchers' and practitioners' attempts to enhance code generation models, retraining and fine-tuning large language models is time-consuming and resource-intensive. Thus, we describe FRANC, a lightweight framework for recommending more secure and high-quality source code derived from transformer-based code generation models. FRANC includes a static filter to make the generated code compilable with heuristics and a quality-aware ranker to sort the code snippets based on a quality score. Moreover, the framework uses prompt engineering to fix persistent quality issues. We evaluated the framework with five Python and Java code generation models and six prompt datasets, including a newly created one in this work (SOEval). The static filter improves 9% to 46% Java suggestions and 10% to 43% Python suggestions regarding compilability. The average improvement over the NDCG@10 score for the ranking system is 0.0763, and the repairing techniques repair the highest 80% of prompts. FRANC takes, on average, 1.98 seconds for Java; for Python, it takes 0.08 seconds.

SciCode is the standard measure of the scientific-coding ability of language models: research-level problems that demand both frontier scientific theory and its implementation as working numerical code. It is a component of the Artificial Analysis Intelligence Index and a standing evaluation in government and national-laboratory suites. Yet its scores have recently plateaued: the strongest 2026 models cluster tightly around 60\% subproblem accuracy, and a successor model ties its predecessor. We trace this stagnation to defects in the benchmark itself. A per-problem, domain-expert audit of all 65 test problems uncovers 263 defects; 192 of them, spread across 91\% of the main problems, cause correct, instruction-following solutions to be wrongly rejected---through non-reproducible gold answers, over-tight tolerances, or self-contradictory specifications. Critically, 78\% of these score-suppressing defects require specialized physics or mathematics knowledge to detect, not mere clerical proofreading. We corrected every confirmable defect to produce SciCode-Verified. The corrections add only the specifications a well-posed problem requires, repair grading, and tighten the tests that were too lenient; every change is recorded with its justification and independently re-checked by a second domain expert. We re-evaluate twelve frontier model snapshots on the corrected benchmark and find a substantial recovery: subproblem accuracy rises from 45--60\% to 84--98\%, and main-problem accuracy from 9--27\% to 69--92\%. State-of-the-art models are far more proficient in scientific coding than SciCode has suggested---the bottleneck was not model capability, but the quality of the evaluation instrument. We release SciCode-Verified with its complete audit trail as the corrected public standard.

This study evaluates the performance of OpenAI's o1-preview model in higher-order cognitive domains, including critical thinking, systematic thinking, computational thinking, data literacy, creative thinking, logical reasoning, and scientific reasoning. Using established benchmarks, we compared the o1-preview models's performance to human participants from diverse educational levels. o1-preview achieved a mean score of 24.33 on the Ennis-Weir Critical Thinking Essay Test (EWCTET), surpassing undergraduate (13.8) and postgraduate (18.39) participants (z = 1.60 and 0.90, respectively). In systematic thinking, it scored 46.1, SD = 4.12 on the Lake Urmia Vignette, significantly outperforming the human mean (20.08, SD = 8.13, z = 3.20). For data literacy, o1-preview scored 8.60, SD = 0.70 on Merk et al.'s "Use Data" dimension, compared to the human post-test mean of 4.17, SD = 2.02 (z = 2.19). On creative thinking tasks, the model achieved originality scores of 2.98, SD = 0.73, higher than the human mean of 1.74 (z = 0.71). In logical reasoning (LogiQA), it outperformed humans with average 90%, SD = 10% accuracy versus 86%, SD = 6.5% (z = 0.62). For scientific reasoning, it achieved near-perfect performance (mean = 0.99, SD = 0.12) on the TOSLS,, exceeding the highest human scores of 0.85, SD = 0.13 (z = 1.78). While o1-preview excelled in structured tasks, it showed limitations in problem-solving and adaptive reasoning. These results demonstrate the potential of AI to complement education in structured assessments but highlight the need for ethical oversight and refinement for broader applications.

In this white paper we introduce Helix, an AI based solution for missense pathogenicity prediction. With recent advances in the sequencing of human genomes, massive amounts of genetic data have become available. This has shifted the burden of labor for genetic diagnostics and research from the gathering of data to its interpretation. Helix presents a state of the art platform for the prediction of pathogenicity in human missense variants. In addition to offering best-in-class predictive performance, Helix offers a platform that allows researchers to analyze and interpret variants in depth that can be accessed at helixlabs.ai.

Self-improving coding agents that iteratively rewrite their own source code have demonstrated impressive performance on coding tasks. However, existing solutions generally derive self-modification from a single failure trajectory at a time, overlooking rich comparative signals available in the agent's expanding archive of past attempts. According to Mendelian principles of controlled inheritance, we introduce Mendel Gödel Machine (MGM). In addition to the general single-trajectory clonal mutation, MGM includes two new types of self-modification that better utilizes evidences accumulated: the reaction-norm mutation edits an agent based on its trajectories on multiple tasks simultaneously, and the cross-lineage hybridization edits an agent using the trajectory of a reference agent from another lineage on the same task. Under an additive fitness landscape model, we prove theoretically and demonstrate via controlled surrogate simulation that the new strategies facilitate a faster and better convergence over single-trajectory baselines. Experiments on SWE-bench and Polyglot confirm MGM's consistent improvement in performance, efficiency, and generalizability.

Agentic systems are evaluated on benchmarks where agents interact with environments to solve tasks. Most papers report a pass@1 score computed from a single run per task, assuming this gives a reliable performance estimate. We test this assumption by collecting 60,000 agentic trajectories on SWE-Bench-Verified, spanning three models and two scaffolds. We find substantial variance: single-run pass@1 estimates vary by 2.2 to 6.0 percentage points depending on which run is selected, with standard deviations exceeding 1.5 percentage points even at temperature 0. This variance has critical implications: reported improvements of 2--3 percentage points may reflect evaluation noise rather than genuine algorithmic progress. Through token-level analysis, we show that trajectories diverge early, often within the first few percent of tokens, and that these small differences cascade into different solution strategies. To enable reliable evaluation of agentic systems, we recommend three concrete practices: (1) estimate pass@1 from multiple independent runs per task, especially when measuring small improvements, (2) use statistical power analysis to determine the number of runs needed to detect expected effect sizes, and (3) consider metrics like pass@k (optimistic bound) and pass^k (pessimistic bound) with k>1 to better characterize the full performance envelope. While these practices increase evaluation cost, they are essential for distinguishing genuine scientific progress from statistical noise.

In this work we design an end-to-end deep learning architecture for predicting, on Chest X-rays images (CXR), a multi-regional score conveying the degree of lung compromise in COVID-19 patients. Such semi-quantitative scoring system, namely Brixia~score, is applied in serial monitoring of such patients, showing significant prognostic value, in one of the hospitals that experienced one of the highest pandemic peaks in Italy. To solve such a challenging visual task, we adopt a weakly supervised learning strategy structured to handle different tasks (segmentation, spatial alignment, and score estimation) trained with a "from-the-part-to-the-whole" procedure involving different datasets. In particular, we exploit a clinical dataset of almost 5,000 CXR annotated images collected in the same hospital. Our BS-Net demonstrates self-attentive behavior and a high degree of accuracy in all processing stages. Through inter-rater agreement tests and a gold standard comparison, we show that our solution outperforms single human annotators in rating accuracy and consistency, thus supporting the possibility of using this tool in contexts of computer-assisted monitoring. Highly resolved (super-pixel level) explainability maps are also generated, with an original technique, to visually help the understanding of the network activity on the lung areas. We also consider other scores proposed in literature and provide a comparison with a recently proposed non-specific approach. We eventually test the performance robustness of our model on an assorted public COVID-19 dataset, for which we also provide Brixia~score annotations, observing good direct generalization and fine-tuning capabilities that highlight the portability of BS-Net in other clinical settings. The CXR dataset along with the source code and the trained model are publicly released for research purposes.

Neural Architecture Search (NAS) has garnered significant research interest due to its capability to discover architectures superior to manually designed ones. Learning text representation is crucial for text classification and other language-related tasks. The NAS model used in text classification does not have a Hybrid hierarchical structure, and there is no restriction on the architecture structure, due to which the search space becomes very large and mostly redundant, so the existing RL models are not able to navigate the search space effectively. Also, doing a flat architecture search leads to an unorganised search space, which is difficult to traverse. For this purpose, we propose HHNAS-AM (Hierarchical Hybrid Neural Architecture Search with Adaptive Mutation Policies), a novel approach that efficiently explores diverse architectural configurations. We introduce a few architectural templates to search on which organise the search spaces, where search spaces are designed on the basis of domain-specific cues. Our method employs mutation strategies that dynamically adapt based on performance feedback from previous iterations using Q-learning, enabling a more effective and accelerated traversal of the search space. The proposed model is fully probabilistic, enabling effective exploration of the search space. We evaluate our approach on the database id (db_id) prediction task, where it consistently discovers high-performing architectures across multiple experiments. On the Spider dataset, our method achieves an 8% improvement in test accuracy over existing baselines.

Whole-genome sequencing (WGS) has revealed numerous non-coding short variants whose functional impacts remain poorly understood. Despite recent advances in deep-learning genomic approaches, accurately predicting and prioritizing clinically relevant mutations in gene regulatory regions remains a major challenge. Here we introduce Deep VRegulome, a deep-learning method for prediction and interpretation of functionally disruptive variants in the human regulome, which combines 700 DNABERT fine-tuned models, trained on vast amounts of ENCODE gene regulatory regions, with variant scoring, motif analysis, attention-based visualization, and survival analysis. We showcase its application on TCGA glioblastoma WGS dataset in prioritizing survival-associated mutations and regulatory regions. The analysis identified 572 splice-disrupting and 9,837 transcription-factor binding site altering mutations occurring in greater than 10% of glioblastoma samples. Survival analysis linked 1352 mutations and 563 disrupted regulatory regions to patient outcomes, enabling stratification via non-coding mutation signatures. All the code, fine-tuned models, and an interactive data portal are publicly available.

Benchmarks for scientific artificial intelligence are mostly written to be scored: multiple-choice questions, curated agent tasks with reference solutions, or simulators with a known generative structure. Real scientific requests arrive differently. They are underspecified, they carry attachments, and lack ground truth. We report K-Bench 01, an evaluation built from first-turn requests sampled from live user traffic on K-Dense Web and run end to end by nine frontier models in identical sandboxes, yielding 1,602 completed agent runs. Three blinded language-model judges scored every run against an eight-dimension rubric. On a rubric whose 8-anchor instructs judges that a domain scientist would accept the work with minor edits, no model clears the line under all three judges. gpt-5.6-sol has the highest pooled mean, 8.04, but its 95% interval [7.80, 8.23] spans the threshold, and two of the three judges rank claude-opus-5 first instead. We therefore report the ordering of systems as the reproducible quantity, the absolute level as an attribute of the instrument, and the top of the table as unresolved. Across all 39,934 scored judgments -- the eight dimension scores plus a holistic overall for each assessment, excluding not-applicable cells -- 47.6% fall below the 8-point threshold. Difficulty is not uniform across the rubric: scientific accuracy averages 6.22 against 7.33 for communication, on identical denominators and in the same direction within every one of the nine models. The single leading failure tag is overclaiming, on 31.4% of assessments. We argue that the informative quantity for scientific agents is not a leaderboard position but the joint distribution of what was delivered, what was claimed, and what artifacts were produced.

Every protein has a unique stability landscape, but the physical consequences of mutation are governed by recurring biochemical constraints. We test whether a shared decoder, trained on measurements from diverse proteins, can interpret these constraints in an unseen target, enabling cross-protein transfer for initial experimental round prescreening. We present StabilityArc , which maps frozen ESMC-600M residue representations through a shared RoPE transformer to an Lx20 matrix of substitution effects; a symmetric, contact-aware residual aids in predicting epistasis in simultaneous substitutions. In 66 strict leave-one-protein-out evaluations covering 134,794 ProteinGym variants, StabilityArc achieves 0.7134 Spearman correlation, exceeding the strongest zero-shot baseline, ProSST-2048 (0.6526), by 0.0608. We further explore the utility of this method by providing the score as a prior for Kermut, achieving Spearman correlation of 0.8280 across three supervised split schemes, improving on Kermut's reported 0.8167.

Many peer-review venues are using algorithms to assign submissions to reviewers. The crux of such automated approaches is the notion of the "similarity score" -- a numerical estimate of the expertise of a reviewer in reviewing a paper -- and many algorithms have been proposed to compute these scores. However, these algorithms have not been subjected to a principled comparison, making it difficult for stakeholders to choose the algorithm in an evidence-based manner. The key challenge in comparing existing algorithms and developing better algorithms is the lack of publicly available gold-standard data. We address this challenge by collecting a novel dataset of similarity scores that we release to the research community. Our dataset consists of 477 self-reported expertise scores provided by 58 researchers who evaluated their expertise in reviewing papers they have read previously. Using our dataset, we compare several widely used similarity algorithms and offer key insights. First, all algorithms exhibit significant error, with misranking rates between 12%-30% in easier cases and 36%-43% in harder ones. Second, most specialized algorithms are designed to work with titles and abstracts of papers, and in this regime the SPECTER2 algorithm performs best. Interestingly, classical TF-IDF matches SPECTER2 in accuracy when given access to full submission texts. In contrast, off-the-shelf LLMs lag behind specialized approaches.

Large language model (LLM) agents are increasingly capable of automating components of machine learning development, yet existing biomedical benchmarks mainly focus on question answering, reasoning, and tool usage, or evaluate only narrow aspects of biomedical ML coding. We present BioXArena, a biomedical machine learning benchmark designed to evaluate whether agents can generate task-specific model training pipelines for heterogeneous and multi-modal biomedical datasets. BioXArena contains 76 end-to-end tasks across 9 domains, including sequence modeling, single-cell analysis, structural biology, network biology, chemical biology, perturbation dynamics, phenotype-disease modeling, biomedical imaging, and text-integrated learning. Each task is curated from primary biomedical sources into a unified evaluation framework with hidden labels, held-out graders, and biology-aware metrics normalized to a 0 to 1 scale. Agents are required to write executable code, train predictive models, and generate submissions for private test samples. Most tasks involve multiple input modalities, including tabular data, images, natural language, molecular sequences, omics matrices, and protein structures. We evaluate 11 agent configurations in a standardized 2-hour single-GPU environment. MLEvolve with Gemini-3.1-Pro achieves the highest average score of 0.666, followed by GPT-5.4 with 0.636, while no single agent consistently dominates across all domains. We additionally perform extensive ablation studies, robustness evaluations, scaling analyses, cost analyses, and failure-mode investigations to better understand how model backbones, agent scaffolds, inference budgets, and biomedical domains influence BioML coding performance. We will publicly release all benchmark tasks, graders, execution runners, leaderboard results, and agent trajectories.

We present Ouroboros, a self-developing agent harness whose tools, prompts, context assembly, and core implementation improve through reviewed commits that become the runtime for later work. Core evolution proceeds in two modes. In recursive free evolution, improvement is itself a task, and completing one evolution cycle can schedule the next. In experience-driven core evolution, ordinary work and social interaction expose bugs, rough edges, and inefficient context construction that lead to reviewed structural changes. On Terminal-Bench 2.1, an Opus 5 run scores 86.74%, the best result reported on the benchmark. On OSWorld-Verified, an Opus 5 run reaches 90.69%, exceeding the best previously reported score. A five-rollout CL-Bench campaign achieves a normalized reward of 0.2301, setting a new state of the art. Hope is the longest-running publicly documented Ouroboros deployment. It is a 161-day living agent experiment in free evolution under governed human communication across seven surfaces. Human interaction surfaces faults and generates proposals, but the agent decides which changes to pursue. Because a self-developing agent may rewrite its own code and select new model APIs, operational safety becomes a primary design problem: guardrails must remain authoritative under evolutionary and public social pressure. Benchmark campaigns use frozen system snapshots, while Hope continues live evolution on a separate lineage.

We introduce a protein language model for determining the complete sequence of a peptide based on measurement of a limited set of amino acids. To date, protein sequencing relies on mass spectrometry, with some novel edman degregation based platforms able to sequence non-native peptides. Current protein sequencing techniques face limitations in accurately identifying all amino acids, hindering comprehensive proteome analysis. Our method simulates partial sequencing data by selectively masking amino acids that are experimentally difficult to identify in protein sequences from the UniRef database. This targeted masking mimics real-world sequencing limitations. We then modify and finetune a ProtBert derived transformer-based model, for a new downstream task predicting these masked residues, providing an approximation of the complete sequence. Evaluating on three bacterial Escherichia species, we achieve per-amino-acid accuracy up to 90.5% when only four amino acids ([KCYM]) are known. Structural assessment using AlphaFold and TM-score validates the biological relevance of our predictions. The model also demonstrates potential for evolutionary analysis through cross-species performance. This integration of simulated experimental constraints with computational predictions offers a promising avenue for enhancing protein sequence analysis, potentially accelerating advancements in proteomics and structural biology by providing a probabilistic reconstruction of the complete protein sequence from limited experimental data.

This paper revisits recent code similarity evaluation metrics, particularly focusing on the application of Abstract Syntax Tree (AST) editing distance in diverse programming languages. In particular, we explore the usefulness of these metrics and compare them to traditional sequence similarity metrics. Our experiments showcase the effectiveness of AST editing distance in capturing intricate code structures, revealing a high correlation with established metrics. Furthermore, we explore the strengths and weaknesses of AST editing distance and prompt-based GPT similarity scores in comparison to BLEU score, execution match, and Jaccard Similarity. We propose, optimize, and publish an adaptable metric that demonstrates effectiveness across all tested languages, representing an enhanced version of Tree Similarity of Edit Distance (TSED).

In recent years, diffusion models trained on equilibrium molecular distributions have proven effective for sampling biomolecules. Beyond direct sampling, the score of such a model can also be used to derive the forces that act on molecular systems. However, while classical diffusion sampling usually recovers the training distribution, the corresponding energy-based interpretation of the learned score is often inconsistent with this distribution, even for low-dimensional toy systems. We trace this inconsistency to inaccuracies of the learned score at very small diffusion timesteps, where the model must capture the correct evolution of the data distribution. In this regime, diffusion models fail to satisfy the Fokker--Planck equation, which governs the evolution of the score. We interpret this deviation as one source of the observed inconsistencies and propose an energy-based diffusion model with a Fokker--Planck-derived regularization term to enforce consistency. We demonstrate our approach by sampling and simulating multiple biomolecular systems, including fast-folding proteins, and by introducing a state-of-the-art transferable Boltzmann emulator for dipeptides that supports simulation and achieves improved consistency and efficient sampling. Our code, model weights, and self-contained JAX and PyTorch notebooks are available at https://github.com/noegroup/ScoreMD.

Residual connections are central to modern deep neural networks, enabling stable optimization and efficient information flow across depth. In this work, we propose SCORE (Skip-Connection ODE Recurrent Embedding), a discrete recurrent alternative to classical layer stacking. Instead of composing multiple independent layers, SCORE iteratively applies a single shared neural block using an ODE (Ordinary Differential Equation)-inspired contractive update: ht+1 = (1 - dt) * ht + dt * F(ht) This formulation can be interpreted as a depth-by-iteration refinement process, where the step size dt explicitly controls stability and update magnitude. Unlike continuous Neural ODE approaches, SCORE uses a fixed number of discrete iterations and standard backpropagation without requiring ODE solvers or adjoint methods. We evaluate SCORE across graph neural networks (ESOL molecular solubility), multilayer perceptrons, and Transformer-based language models (nanoGPT). Across architectures, SCORE generally improves convergence speed and often accelerates training. SCORE is reducing parameter count through shared weights. In practice, simple Euler integration provides the best trade-off between computational cost and performance, while higher-order integrators yield marginal gains at increased compute. These results suggest that controlled recurrent depth with contractive residual updates offers a lightweight and effective alternative to classical stacking in deep neural networks.

Mutation-based fuzzing is effective for uncovering compiler bugs, but designing high-quality mutators for modern languages with complex constructs (e.g., templates, macros) remains challenging. Existing methods rely heavily on manual design or human-in-the-loop correction, limiting scalability and cross-language generalizability. We present Mut4All, a fully automated, language-agnostic framework that synthesizes mutators using Large Language Models (LLMs) and compiler-specific knowledge from bug reports. It consists of three agents: (1) a mutator invention agent that identifies mutation targets and generates mutator metadata using compiler-related insights; (2) a mutator implementation synthesis agent, fine-tuned to produce initial implementations; and (3) a mutator refinement agent that verifies and corrects the mutators via unit-test feedback. Mut4All processes 1000 bug reports (500 Rust, 500 C++), yielding 319 Rust and 403 C++ mutators at ~$0.08 each via GPT-4o. Our customized fuzzer, using these mutators, finds 62 bugs in Rust compilers (38 new, 7 fixed) and 34 bugs in C++ compilers (16 new, 1 fixed). Mut4All outperforms existing methods in both unique crash detection and coverage, ranking first on Rust and second on C++.

Despite rapid progress in LLM agents, performance on long-horizon, tool-using tasks remains fragile. To better understand this fragility, we ask a simple question: do all actions contribute equally to failure? Analyzing execution traces on τ-Bench (Airline/Retail) and SWE-Bench Verified, we decompose trajectories into mutating (environment-changing) vs.\ non-mutating steps and formalize decisive deviations, earliest action, level divergences that flip success to failure. A logistic regression reveals that each additional deviation in a mutating action reduces the odds of success by upto 92% on Airline and upto 96% on Retail for SoTA models. In contrast, deviations in non-mutating actions have little to no effect. Errors also grow with context length as agents drift from role and act on stale constraints. Motivated by these observations, we introduce , a model-agnostic, gradient-free, test-time safeguard that (i) adds mutation-gated verification, (ii) injects Targeted Reflection before mutating steps, and (iii) performs block-based context cleaning. delivers consistent gains, e.g., Qwen3-Thinking: +28\% relative on Airline, +11\% on Retail, and +7\% on SWE-Bench Verified; Claude: +9\%/+7\%. We further identify ceiling effects in τ-Bench, where annotation errors and underspecified tasks artificially cap model performance. To address this, we release τ-Bench Verified, which restores benchmark headroom through targeted revisions. Our results argue for action-level analysis, targeted safeguards, and reliable evaluations as prerequisites for robust multi-turn agents.

Medical coding translates clinical documentation into standardized codes for billing, research, and public health, but manual coding is time-consuming and error-prone. Existing automation efforts rely on small datasets that poorly represent real-world patient heterogeneity. We trained a language model on 5.8 million electronic health records from 1.8 million patients across nearly all specialties in Eastern Denmark (2006--2016) to predict ICD-10 codes from clinical notes, medications, and laboratory results. Evaluated on 270,000 held-out patients, the model achieved a micro F1 of 71.8% and a top-10 recall of 95.5%. Performance varied by specialty (F1: 53--91%), with higher scores in specialties with well-defined diagnostic criteria. Codes appearing predominantly as secondary diagnoses had markedly lower F1 scores. For three such codes (suicide-related behaviors, weight disorders, and hypertension), the model identified thousands of uncoded cases, of which 76-86% were confirmed valid upon manual review, suggesting systematic under-coding rather than model error. These findings suggest under-coding of secondary diagnoses in Eastern Denmark during this period, with potential implications for epidemiological research, public health surveillance, and understanding of multimorbidity. Similar time constraints and reimbursement structures in other healthcare systems suggest this may not be isolated to this dataset. The model can automate coding for approximately 50% of cases and provide accurate suggestions for most others, and may offer a practical solution to help capture missed secondary conditions.

Benchmarks for LLM-generated GPU kernels decide correctness with a few random inputs and a loose floating-point tolerance, and their verdicts now feed leaderboards and reinforcement-learning rewards. Recent work agrees these checkers are weak and patches them by hand---extra input distributions, fuzzing recipes, tighter tolerances---with no way to measure whether any patch suffices. We introduce mutation analysis as an adequacy metric for kernel-benchmark oracles: deterministic rules inject 10{,}303 compilable faults into verified CUDA implementations of 188 KernelBench problems, 7{,}384 of them with an independent kill witness; any test protocol is scored by the fraction it detects. The official check misses one in six witnessed faults (16.9%), deterministically, and the misses are skewed by family: 8.7% of arithmetic faults escape, but 78.6% of precision faults do. The metric explains why (a tolerance blind band growing with reduction size; a measured ceiling on input aggressiveness set by legitimate floating-point variance), audits the strongest existing patch (KernelBench-Verified's gain splits into +4.0 points from hidden inputs and +4.5 from tighter tolerance, a split its authors could not compute), and exposes a published fuzzing recipe that rejects correct kernels 107 times. Optimizing suites over the kill matrix reaches 98.0% detection with two inputs per problem (94.8% held-out), and the measurement's fault taxonomy teaches a test generator more than the raw faults themselves. Across 48 whole architectures, the blindness grows with scale, concentrating in deep homogeneous pipelines, and two problems prove unrefereeable: their official references violate the benchmark's own tolerance against fp64. We release everything as https://huggingface.co/datasets/Elfsong/KernelBench-M{KernelBench-M}.

Diversity is an important criterion for many areas of machine learning (ML), including generative modeling and dataset curation. Yet little work has gone into understanding, formalizing, and measuring diversity in ML. In this paper, we address the diversity evaluation problem by proposing the Vendi Score, which connects and extends ideas from ecology and quantum statistical mechanics to ML. The Vendi Score is defined as the exponential of the Shannon entropy of the eigenvalues of a similarity matrix. This matrix is induced by a user-defined similarity function applied to the sample to be evaluated for diversity. In taking a similarity function as input, the Vendi Score enables its user to specify any desired form of diversity. Importantly, unlike many existing metrics in ML, the Vendi Score doesn't require a reference dataset or distribution over samples or labels, it is therefore general and applicable to any generative model, decoding algorithm, and dataset from any domain where similarity can be defined. We showcased the Vendi Score on molecular generative modeling, a domain where diversity plays an important role in enabling the discovery of novel molecules. We found that the Vendi Score addresses shortcomings of the current diversity metric of choice in that domain. We also applied the Vendi Score to generative models of images and decoding algorithms of text and found it confirms known results about diversity in those domains. Furthermore, we used the Vendi Score to measure mode collapse, a known limitation of generative adversarial networks (GANs). In particular, the Vendi Score revealed that even GANs that capture all the modes of a labeled dataset can be less diverse than the original dataset. Finally, the interpretability of the Vendi Score allowed us to diagnose several benchmark ML datasets for diversity, opening the door for diversity-informed data augmentation.

The molecular characterization of tumor samples by multiple omics data sets of different types or modalities (e.g. gene expression, mutation, CpG methylation) has become an invaluable source of information for assessing the expected performance of individual drugs and their combinations. Merging the relevant information from the omics data modalities provides the statistical basis for determining suitable therapies for specific cancer patients. Different data modalities may each have their specific structures that need to be taken into account during inference. In this paper, we assume that each omics data modality has a low-rank structure with only few relevant features that affect the prediction and we propose to use a composite local nuclear norm penalization for learning drug sensitivity. Numerical results show that the composite low-rank structure can improve the prediction performance compared to using a global low-rank approach or elastic net regression.

Information tasks such as writing surveys or analytical reports require complex search and reasoning, and have recently been grouped under the umbrella of deep research -- a term also adopted by recent models targeting these capabilities. Despite growing interest, the scope of the deep research task remains underdefined and its distinction from other reasoning-intensive problems is poorly understood. In this paper, we propose a formal characterization of the deep research (DR) task and introduce a benchmark to evaluate the performance of DR systems. We argue that the core defining feature of deep research is not the production of lengthy report-style outputs, but rather the high fan-out over concepts required during the search process, i.e., broad and reasoning-intensive exploration. To enable objective evaluation, we define DR using an intermediate output representation that encodes key claims uncovered during search-separating the reasoning challenge from surface-level report generation. Based on this formulation, we propose a diverse, challenging benchmark LiveDRBench with 100 challenging tasks over scientific topics (e.g., datasets, materials discovery, prior art search) and public interest events (e.g., flight incidents, movie awards). Across state-of-the-art DR systems, F1 score ranges between 0.02 and 0.72 for any sub-category. OpenAI's model performs the best with an overall F1 score of 0.55. Analysis of reasoning traces reveals the distribution over the number of referenced sources, branching, and backtracking events executed by current DR systems, motivating future directions for improving their search mechanisms and grounding capabilities. The benchmark is available at https://github.com/microsoft/LiveDRBench.

Symbolic regression (SR) is the problem of learning a symbolic expression from numerical data. Recently, deep neural models trained on procedurally-generated synthetic datasets showed competitive performance compared to more classical Genetic Programming (GP) algorithms. Unlike their GP counterparts, these neural approaches are trained to generate expressions from datasets given as context. This allows them to produce accurate expressions in a single forward pass at test time. However, they usually do not benefit from search abilities, which result in low performance compared to GP on out-of-distribution datasets. In this paper, we propose a novel method which provides the best of both worlds, based on a Monte-Carlo Tree Search procedure using a context-aware neural mutation model, which is initially pre-trained to learn promising mutations, and further refined from successful experiences in an online fashion. The approach demonstrates state-of-the-art performance on the well-known SRBench benchmark.

Foundation models are increasingly used in scientific research, but evaluating AI-generated scientific work remains challenging. While expert reviews are costly, large language models (LLMs) as proxy reviewers have proven to be unreliable. To address this, we investigate two automatic evaluation metrics, specifically citation count prediction and review score prediction. We parse all papers of OpenReview and augment each submission with its citation count, reference, and research hypothesis. Our findings reveal that citation count prediction is more viable than review score prediction, and predicting scores is more difficult purely from the research hypothesis than from the full paper. Furthermore, we show that a simple prediction model based solely on title and abstract outperforms LLM-based reviewers, though it still falls short of human-level consistency.

LLM agents often operate over multi-turn interactions in which user intent changes before execution. We study intent drift: the failure mode in which superseded parts of the user's intent continue to influence the final answer or tool action. We introduce IntentFlux, an executable benchmark that converts verifiable tasks into dialogues with controlled intent changes while preserving their original graders. In a 627-case calibration, mean task score falls from 0.476 to 0.384 as dialogues contain more superseded and withdrawn information. Across eight models, the rate of fully correct solutions is significantly lower when the same final task must be recovered from an evolving dialogue rather than given directly in a single turn. We further introduce StateForge, which explicitly maintains the active requirements before generation. On General-Test, it improves mean task score from 0.367 to 0.467. Providing the ground-truth final state improves performance further but still does not recover single-turn performance, indicating that state-estimation errors explain only part of the gap. These results establish intent drift as a measurable multi-turn failure mode and explicit state maintenance as a partial mitigation.

Recent advancements in language modeling have enabled the translation of natural language into code, and the use of execution feedback to improve code generation. However, these methods often rely heavily on pre-existing test cases, which may not always be available or comprehensive. In this work, we propose a novel approach that concurrently trains a code generation model and a test generation model, utilizing execution feedback to refine and enhance the performance of both. We introduce two strategies for test and code data augmentation and a new scoring function for code and test ranking. We experiment on the APPS dataset and demonstrate that our approach can effectively generate and augment test cases, filter and synthesize correct code solutions, and rank the quality of generated code and tests. The results demonstrate that our models, when iteratively trained with an increasing number of test cases and code solutions, outperform those trained on the original dataset.

AI agents are increasingly deployed in production, yet their security evaluations remain bottlenecked by manual red-teaming or static benchmarks that fail to model adaptive, multi-turn adversaries. We propose NAAMSE, an evolutionary framework that reframes agent security evaluation as a feedback-driven optimization problem. Our system employs a single autonomous agent that orchestrates a lifecycle of genetic prompt mutation, hierarchical corpus exploration, and asymmetric behavioral scoring. By using model responses as a fitness signal, the framework iteratively compounds effective attack strategies while simultaneously ensuring "benign-use correctness", preventing the degenerate security of blanket refusal. Our experiments on Gemini 2.5 Flash demonstrate that evolutionary mutation systematically amplifies vulnerabilities missed by one-shot methods, with controlled ablations revealing that the synergy between exploration and targeted mutation uncovers high-severity failure modes. We show that this adaptive approach provides a more realistic and scalable assessment of agent robustness in the face of evolving threats. The code for NAAMSE is open source and available at https://github.com/HASHIRU-AI/NAAMSE.

It is well known that anti-malware scanners depend on malware signatures to identify malware. However, even minor modifications to malware code structure results in a change in the malware signature thus enabling the variant to evade detection by scanners. Therefore, there exists the need for a proactively generated malware variant dataset to aid detection of such diverse variants by automated antivirus scanners. This paper proposes and demonstrates a generic assembly source code based framework that facilitates any evolutionary algorithm to generate diverse and potential variants of an input malware, while retaining its maliciousness, yet capable of evading antivirus scanners. Generic code transformation functions and a novelty search supported quality metric have been proposed as components of the framework to be used respectively as variation operators and fitness function, for evolutionary algorithms. The results demonstrate the effectiveness of the framework in generating diverse variants and the generated variants have been shown to evade over 98% of popular antivirus scanners. The malware variants evolved by the framework can serve as antigens to assist malware analysis engines to improve their malware detection algorithms.

Multi-turn-to-single-turn (M2S) compresses iterative red-teaming into one structured prompt, but prior work relied on a handful of manually written templates. We present X-Teaming Evolutionary M2S, an automated framework that discovers and optimizes M2S templates through language-model-guided evolution. The system pairs smart sampling from 12 sources with an LLM-as-judge inspired by StrongREJECT and records fully auditable logs. Maintaining selection pressure by setting the success threshold to theta = 0.70, we obtain five evolutionary generations, two new template families, and 44.8% overall success (103/230) on GPT-4.1. A balanced cross-model panel of 2,500 trials (judge fixed) shows that structural gains transfer but vary by target; two models score zero at the same threshold. We also find a positive coupling between prompt length and score, motivating length-aware judging. Our results demonstrate that structure-level search is a reproducible route to stronger single-turn probes and underscore the importance of threshold calibration and cross-model evaluation. Code, configurations, and artifacts are available at https://github.com/hyunjun1121/M2S-x-teaming.

We introduce BM25S, an efficient Python-based implementation of BM25 that only depends on Numpy and Scipy. BM25S achieves up to a 500x speedup compared to the most popular Python-based framework by eagerly computing BM25 scores during indexing and storing them into sparse matrices. It also achieves considerable speedups compared to highly optimized Java-based implementations, which are used by popular commercial products. Finally, BM25S reproduces the exact implementation of five BM25 variants based on Kamphuis et al. (2020) by extending eager scoring to non-sparse variants using a novel score shifting method. The code can be found at https://github.com/xhluca/bm25s

Accurate interpretation of lab results is crucial in clinical medicine, yet most patient portals use universal normal ranges, ignoring conditional factors like age and gender. This study introduces Lab-AI, an interactive system that offers personalized normal ranges using retrieval-augmented generation (RAG) from credible health sources. Lab-AI has two modules: factor retrieval and normal range retrieval. We tested these on 122 lab tests: 40 with conditional factors and 82 without. For tests with factors, normal ranges depend on patient-specific information. Our results show GPT-4-turbo with RAG achieved a 0.948 F1 score for factor retrieval and 0.995 accuracy for normal range retrieval. GPT-4-turbo with RAG outperformed the best non-RAG system by 33.5% in factor retrieval and showed 132% and 100% improvements in question-level and lab-level performance, respectively, for normal range retrieval. These findings highlight Lab-AI's potential to enhance patient understanding of lab results.

LLMs have become the go-to choice for code generation tasks, with an exponential increase in the training, development, and usage of LLMs specifically for code generation. To evaluate the ability of LLMs on code, both academic and industry practitioners rely on popular handcrafted benchmarks. However, prior benchmarks contain only a very limited set of problems, both in quantity and variety. Further, due to popularity and age, many benchmarks are prone to data leakage where example solutions can be readily found on the web and thus potentially in training data. Such limitations inevitably lead us to inquire: Is the leaderboard performance on existing benchmarks reliable and comprehensive enough to measure the program synthesis ability of LLMs? To address this, we introduce EvoEval -- a program synthesis benchmark suite created by evolving existing benchmarks into different targeted domains for a comprehensive evaluation of LLM coding abilities. Our study on 51 LLMs shows that compared to the high performance obtained on standard benchmarks like HumanEval, there is a significant drop in performance (on average 39.4%) when using EvoEval. Additionally, the decrease in performance can range from 19.6% to 47.7%, leading to drastic ranking changes amongst LLMs and showing potential overfitting of existing benchmarks. Furthermore, we showcase various insights, including the brittleness of instruction-following models when encountering rewording or subtle changes as well as the importance of learning problem composition and decomposition. EvoEval not only provides comprehensive benchmarks, but can be used to further evolve arbitrary problems to keep up with advances and the ever-changing landscape of LLMs for code. We have open-sourced our benchmarks, tools, and complete LLM generations at https://github.com/evo-eval/evoeval

Language models have emerged as powerful predictors of the viability of biological sequences. During training these models learn the rules of the grammar obeyed by sequences of amino acids or nucleotides. Once trained, these models can take a sequence as input and produce a likelihood score as an output; a higher likelihood implies adherence to the learned grammar and correlates with experimental fitness measurements. Here we show that in-context learning can distort the relationship between fitness and likelihood scores of sequences. This phenomenon most prominently manifests as anomalously high likelihood scores for sequences that contain repeated motifs. We use protein language models with different architectures trained on the masked language modeling objective for our experiments, and find transformer-based models to be particularly vulnerable to this effect. This behavior is mediated by a look-up operation where the model seeks the identity of the masked position by using the other copy of the repeated motif as a reference. This retrieval behavior can override the model's learned priors. This phenomenon persists for imperfectly repeated sequences, and extends to other kinds of biologically relevant features such as reversed complement motifs in RNA sequences that fold into hairpin structures.

Biological reasoning models use post-training to connect LLMs to biological foundation model representations and biological text. Their benchmark accuracy is taken as evidence that LLMs reason over these inputs. We test this assumption in six biological reasoning models across DNA, protein, and single-cell tasks. We perturb one biological input while holding the query and other inputs fixed, construct evidence conflicts that pair the foundation model representation of one genome, protein, or cell with the text of another, fit linear probes to the representations the language model receives, and analyze reasoning traces against the biological inputs. Evo2 and ESM3 contribute little to BioReason and BioReason-Pro performance on the evaluated tasks. Shuffling the DNA sequence barely changes BioReason disease prediction accuracy, and in evidence conflicts the two models follow the text in 97.9% and 99.7% of cases. Linear probes trained on the Evo2 and ESM3 representations predict the task targets, so these foundation models encode information relevant to the task, but provide limited overall performance improvement to BioReason and BioReason-Pro. In contrast, foundation model inputs contribute to ChatNT, Prot2Text-V2, and CellWhisperer performance, and differentially expressed genes in the gene sentence contribute to Cell2Sentence-Scale performance. Across SFT and RL checkpoints of BioReason-Pro and 42 BioReason checkpoints, increases in accuracy do not imply greater performance contributions from biological inputs. BioReason traces misstate nucleotide changes, while BioReason-Pro traces describe functions omitted from final predictions under evidence conflicts. We find that current post-training strategies do not ensure that foundation model representations contribute to task performance.

Recent advances in large language models for test case generation have improved branch coverage via prompt-engineered mutations. However, they still lack principled mechanisms for steering models toward specific high-risk execution branches, limiting their effectiveness for discovering subtle bugs and security vulnerabilities. We propose GLMTest, the first program structure-aware LLM framework for targeted test case generation that seamlessly integrates code property graphs and code semantics using a graph neural network and a language model to condition test case generation on execution branches. This structured conditioning enables controllable and branch-targeted test case generation, thereby potentially enhancing bug and security risk discovery. Experiments on real-world projects show that GLMTest built on a Qwen2.5-Coder-7B-Instruct model improves branch accuracy from 27.4% to 50.2% on TestGenEval benchmark compared with state-of-the-art LLMs, i.e., Claude-Sonnet-4.5 and GPT-4o-mini.

A benchmark score compares a system output against a reference, and methodological attention falls almost entirely on the first term. We measure the second. The retained annotation record of a six-language benchmark for personally identifiable information contains two independent annotator labellings, the aggregate shipped as gold, and a reviewer gold from independent expert re-annotation of a sample. Using it, we hold the scored output fixed and exchange the reference. The score moves by 4.95 F1 points for one output and 2.00 for the other (95% CIs [3.23, 6.27] and [0.26, 3.37]), and by 7.55 in the worst language. The comparison between two outputs moves as well: the paired interaction between reference and system is +2.95 points (CI [+1.97, +3.87]), survives correction for multiple testing, and changes one language's margin outright. The cause is an undocumented aggregation default that usually kept one annotator when adjudication did not fire, making the shipped reference a partial copy of an output being scored. We report the resulting reference-sensitivity band, show how to compute one from any retained annotation record, and argue that the quantity belongs beside the score.

This paper presents the RMIT--ADM+S participation in the SIGIR 2025 LiveRAG Challenge. Our Generation-Retrieval-Augmented Generation (GRAG) approach relies on generating a hypothetical answer that is used in the retrieval phase, alongside the original question. GRAG also incorporates a pointwise large language model (LLM)-based re-ranking step prior to final answer generation. We describe the system architecture and the rationale behind our design choices. In particular, a systematic evaluation using the Grid of Points (GoP) framework and N-way ANOVA enabled comparison across multiple configurations, including query variant generation, question decomposition, rank fusion strategies, and prompting techniques for answer generation. Our system achieved a Relevance score of 1.199 and a Faithfulness score of 0.477 on the private leaderboard, placing among the top four finalists in the LiveRAG 2025 Challenge.

Predicting cellular responses to genetic perturbations helps prioritize experiments in single-cell genomics, where exhaustive measurement is infeasible. While computational models increasingly predict these responses, three evaluation deficiencies obscure what their scores demonstrate. First, absolute metrics cannot separate target-specific predictions from a shared background response. Second, common metrics remain high under gene shuffling, so gene-level accuracy is never verified. Third, a score at one training size says nothing about coverage, which depends on representation-space proximity and response-constraining power. We propose AmbiModBench, a specificity-aware, gene-resolved and coverage-aware benchmark. It pairs every score with a training-mean reference fitted on the same split, screens each readout by gene-coordinate permutation, and links embedding distance to response variation. Across K562, RPE1 and Norman, strong absolute scores largely reflect shared background rather than target-specific learning. Widely used readouts track response magnitude distributions rather than the affected genes. Detectable gain follows representation-space coverage rather than training-set size. Nonetheless, on RPE1 the protocol yields a reproducible target-specific gain across five additional splits and three gene selections, which absolute scores alone cannot distinguish from shared background.

Program synthesis has been long studied with recent approaches focused on directly using the power of Large Language Models (LLMs) to generate code. Programming benchmarks, with curated synthesis problems and test-cases, are used to measure the performance of various LLMs on code synthesis. However, these test-cases can be limited in both quantity and quality for fully assessing the functional correctness of the generated code. Such limitation in the existing benchmarks begs the following question: In the era of LLMs, is the code generated really correct? To answer this, we propose EvalPlus -- a code synthesis evaluation framework to rigorously benchmark the functional correctness of LLM-synthesized code. EvalPlus augments a given evaluation dataset with large amounts of test-cases newly produced by an automatic test input generator, powered by both LLM- and mutation-based strategies. While EvalPlus is general, we extend the test-cases of the popular HumanEval benchmark by 80x to build HumanEval+. Our extensive evaluation across 26 popular LLMs (e.g., GPT-4 and ChatGPT) demonstrates that HumanEval+ is able to catch significant amounts of previously undetected wrong code synthesized by LLMs, reducing the pass@k by up-to 19.3-28.9%. We also surprisingly found that test insufficiency can lead to mis-ranking. For example, both WizardCoder-CodeLlama and Phind-CodeLlama now outperform ChatGPT on HumanEval+, while none of them could on HumanEval. Our work not only indicates that prior popular code synthesis evaluation results do not accurately reflect the true performance of LLMs for code synthesis, but also opens up a new direction to improve such programming benchmarks through automated testing. We have open-sourced our tools, enhanced datasets as well as all LLM-generated code at https://github.com/evalplus/evalplus to facilitate and accelerate future LLM-for-code research.

Agent benchmarks increasingly evaluate repository editing, web research, terminal use, and long-horizon interaction. Their scores support capability claims only when the evaluation protocol keeps the intended capability necessary for success. Recent reward-hacking benchmarks and system reports show that agents can instead recover public solutions, read evaluation artifacts, infer generator structure, manipulate feedback, or benefit from invalid scoring paths; existing responses do not provide a common procedure for attributing these shortcuts and quantifying their effect across benchmarks. We formulate protocol validity and introduce HackDetect, a post-hoc audit that identifies an exposure, determines how the agent used it, and assesses whether the resulting score is misleading. We quantify score inflation with the Mislead gap, defined as the exploit score minus the intended score. We audit 2,385 traces across 15 agent benchmarks and find evidence of exposures and reward hacking in 67.0% of Frontier Science traces and 66.7% of AutoLab tasks. Across paired comparisons, we measure score inflation of 0.45-1.00, showing that benchmark reports should provide evidence that scores reflect the intended capability.

A code generation model generates code by taking a prompt from a code comment, existing code, or a combination of both. Although code generation models (e.g. GitHub Copilot) are increasingly being adopted in practice, it is unclear whether they can successfully be used for unit test generation without fine-tuning. We investigated how well three generative models (Codex, GPT-3.5-Turbo, and StarCoder) can generate test cases to fill this gap. We used two benchmarks (HumanEval and Evosuite SF110) to investigate the context generation's effect in the unit test generation process. We evaluated the models based on compilation rates, test correctness, coverage, and test smells. We found that the Codex model achieved above 80% coverage for the HumanEval dataset, but no model had more than 2% coverage for the EvoSuite SF110 benchmark. The generated tests also suffered from test smells, such as Duplicated Asserts and Empty Tests.

Multi-modal generative document parsing systems challenge traditional evaluation: unlike deterministic OCR or layout models, they often produce semantically correct yet structurally divergent outputs. Conventional metrics-CER, WER, IoU, or TEDS-misclassify such diversity as error, penalizing valid interpretations and obscuring system behavior. We introduce SCORE (Structural and COntent Robust Evaluation), an interpretation-agnostic framework that integrates (i) adjusted edit distance for robust content fidelity, (ii) token-level diagnostics to distinguish hallucinations from omissions, (iii) table evaluation with spatial tolerance and semantic alignment, and (iv) hierarchy-aware consistency checks. Together, these dimensions enable evaluation that embraces representational diversity while enforcing semantic rigor. Across 1,114 pages spanning a holistic benchmark and a field dataset, SCORE consistently revealed cross-dataset performance patterns missed by standard metrics. In 2-5% of pages with ambiguous table structures, traditional metrics penalized systems by 12-25% on average, leading to distorted rankings. SCORE corrected these cases, recovering equivalence between alternative but valid interpretations. Moreover, by normalizing generative outputs into a format-agnostic representation, SCORE reproduces traditional scores (e.g., table F1 up to 0.93) without requiring object-detection pipelines, demonstrating that generative parsing alone suffices for comprehensive evaluation. By exposing how interpretive diversity impacts evaluation outcomes and providing multi-dimensional, interpretable diagnostics, SCORE establishes foundational principles for semantically grounded, fair, and practical benchmarking of modern document parsing systems.

Self-evolving skill systems promise to improve agents by turning execution feedback into persistent skill updates without changing the underlying model. Yet it remains unclear when further evolution helps, how successful and failed trajectories shape revision, and whether extra test-time computation can recover the same gains. To address these questions, we present a controlled evaluation framework across five benchmarks and three models. Our primary study contains 42 feedback runs across 14 supported model-benchmark settings. Within each setting, we hold the executor and optimizer configuration, revision procedure, validation rule, and round budget fixed, while varying only the feedback shown to the optimizer: successes and failures (Normal), failures only, or successes only. Evolution is sparse: only 55 of 388 candidates establish byte-distinct validation bests. Validation-based selection chooses an evolved skill in 11 of 14 settings, nine of which improve released-test performance. All 11 selections come from feedback conditions that include failed trajectories, although the relative ranking of Normal and Fail-only varies across settings. Validation and downstream evaluations on test, robustness, and transfer sometimes favor different feedback views. A broader SearchQA analysis covering eight models shows similarly sparse, feedback-dependent dynamics. In the GPT-5.5 test-time-scaling controls, oracle Parallel Sampling comes within 0.43 points of the evolved SearchQA skill but remains 30.96 points behind on SpreadsheetBench; Sequential Refinement recovers neither gain. Overall, persistent skill self-evolution is better understood as sparse, validation-filtered search with model- and benchmark-dependent returns, rather than steady improvement from additional rounds. The implementation is available at https://github.com/HKUST-KnowComp/rethinkskill.

Deep research (DR) agents address open-ended information needs through multi-step reasoning, long-horizon web search, and evidence synthesis. However, existing benchmarks primarily focus on academic research or synthetic tasks, leaving DR agents' capabilities in real-world life and professional productivity settings insufficiently evaluated. To bridge this gap, we introduce LifeSearch, a challenging benchmark comprising 203 Chinese and English tasks derived from real-user query seeds across 7 concrete life and productivity domains. With more than 4,500 hours of human annotation, we construct over 4,000 expert-written, fine-grained rubrics, covering both report-content criteria for coverage and accuracy and concrete criteria specifying values, units, dates, and URLs. We conduct extensive experiments on 12 advanced DR agents under a unified evaluation harness, systematically analyzing their performance patterns and failure modes. We find that even the best-performing frontier agent, GPT-5.6 Sol, achieves a score of only 63.4%, leaving substantial room for improvement on realistic life and professional DR tasks. We select the grading configuration by comparing candidate LLM judges against human rubric labels, and a human ranking study shows strong alignment between aggregate benchmark scores and human judgments (Spearman's ρ= 0.89, Kendall's τ= 0.81). LifeSearch provides a rigorous evaluation framework and foundation for advancing DR agents toward real-world life and productivity.

ARC-AGI-3 evaluates agents in interactive environments whose rules and objectives must be inferred from observation. We present Kepler, an open-source harness that represents hypotheses as executable world models and validates them through retrospective transition checks and conditional prediction checks. Under one frozen Claude Opus 5 configuration, Kepler obtained a server-verified 100.00 RHAE on all 25 public games, with no per-game model selection or score-conditioned reruns. On 181 of 183 completed levels, the final Opus attempt used no more actions than the corresponding median-human baseline. The retained board runs used 8,256 environment actions, of which 7,292 occurred in scored levels. Retained local provider-session records yield 858.0 million tokens, 97.37% cache reads, and a \$777.72 cost at September 1, 2026 API list-equivalent rates. We also report three evaluation failures: source-code leakage that produced an invalid perfect run, agents reconstructing a removed harness in a control condition, and autonomous repair masking a broken planner. A single-game observation case study showed that animation frames contained task-relevant information absent from settled text grids. Across the final Claude Opus 5 and GPT-5.6 Sol boards, 48 of 50 game-model cells reached 100. These results indicate that public-set score alone has limited discriminative value and motivate first-attempt, cost-conditioned, and verification-aware reporting.

Passage ranking involves two stages: passage retrieval and passage re-ranking, which are important and challenging topics for both academics and industries in the area of Information Retrieval (IR). However, the commonly-used datasets for passage ranking usually focus on the English language. For non-English scenarios, such as Chinese, the existing datasets are limited in terms of data scale, fine-grained relevance annotation and false negative issues. To address this problem, we introduce T2Ranking, a large-scale Chinese benchmark for passage ranking. T2Ranking comprises more than 300K queries and over 2M unique passages from real-world search engines. Expert annotators are recruited to provide 4-level graded relevance scores (fine-grained) for query-passage pairs instead of binary relevance judgments (coarse-grained). To ease the false negative issues, more passages with higher diversities are considered when performing relevance annotations, especially in the test set, to ensure a more accurate evaluation. Apart from the textual query and passage data, other auxiliary resources are also provided, such as query types and XML files of documents which passages are generated from, to facilitate further studies. To evaluate the dataset, commonly used ranking models are implemented and tested on T2Ranking as baselines. The experimental results show that T2Ranking is challenging and there is still scope for improvement. The full data and all codes are available at https://github.com/THUIR/T2Ranking/

In this paper, we carry out experimental research on Grammatical Error Correction, delving into the nuances of single-model systems, comparing the efficiency of ensembling and ranking methods, and exploring the application of large language models to GEC as single-model systems, as parts of ensembles, and as ranking methods. We set new state-of-the-art performance with F_0.5 scores of 72.8 on CoNLL-2014-test and 81.4 on BEA-test, respectively. To support further advancements in GEC and ensure the reproducibility of our research, we make our code, trained models, and systems' outputs publicly available.

Autonomous research agents produce competitive solutions and professional-looking manuscripts, yet their outputs contain verifiability failures undetectable by surface-level evaluation: fabricated citations, unreproducible scores, and method descriptions that diverge from the implementation. We address this through three contributions. First, Chain-of-Evidence (CoE), a verifiability framework requiring every claim to be traceable to its evidence source. Second, ScientistOne, an end-to-end autonomous research system that maintains evidence chains by construction throughout literature review, solution discovery, and paper writing. Third, CoE Audit, a post-hoc audit whose four integrity checks -- score verification, specification violation, reference verification, and method-code alignment -- apply uniformly to all systems. Across 75 papers spanning five systems and five frontier research tasks, every baseline exhibits at least one systematic failure mode: hallucinated reference rates reach 21%, score verification passes in as few as 42% of papers, and method-code alignment ranges from 20% to 80%. ScientistOne achieves zero hallucinated references (0/337), perfect score verification (12/12), and the highest method-code alignment (14/15), while matching or exceeding human expert performance on all five tasks. ScientistOne further generalizes to six additional tasks spanning medical imaging, fine-grained recognition, 3D perception, and language modeling, achieving state-of-the-art on Parameter Golf and gold medals on MLE-Bench tasks where baselines fail entirely.

We present the findings of "The Alzheimer's Disease Prediction Of Longitudinal Evolution" (TADPOLE) Challenge, which compared the performance of 92 algorithms from 33 international teams at predicting the future trajectory of 219 individuals at risk of Alzheimer's disease. Challenge participants were required to make a prediction, for each month of a 5-year future time period, of three key outcomes: clinical diagnosis, Alzheimer's Disease Assessment Scale Cognitive Subdomain (ADAS-Cog13), and total volume of the ventricles. The methods used by challenge participants included multivariate linear regression, machine learning methods such as support vector machines and deep neural networks, as well as disease progression models. No single submission was best at predicting all three outcomes. For clinical diagnosis and ventricle volume prediction, the best algorithms strongly outperform simple baselines in predictive ability. However, for ADAS-Cog13 no single submitted prediction method was significantly better than random guesswork. Two ensemble methods based on taking the mean and median over all predictions, obtained top scores on almost all tasks. Better than average performance at diagnosis prediction was generally associated with the additional inclusion of features from cerebrospinal fluid (CSF) samples and diffusion tensor imaging (DTI). On the other hand, better performance at ventricle volume prediction was associated with inclusion of summary statistics, such as the slope or maxima/minima of biomarkers. TADPOLE's unique results suggest that current prediction algorithms provide sufficient accuracy to exploit biomarkers related to clinical diagnosis and ventricle volume, for cohort refinement in clinical trials for Alzheimer's disease. However, results call into question the usage of cognitive test scores for patient selection and as a primary endpoint in clinical trials.

Instruction tuning is critical for adapting large language models (LLMs) to downstream tasks, and recent studies have demonstrated that small amounts of human-curated data can outperform larger datasets, challenging traditional data scaling laws. While LLM-based data quality rating systems offer a cost-effective alternative to human annotation, they often suffer from inaccuracies and biases, even in powerful models like GPT-4. In this work, we introduce DS2, a Diversity-aware Score curation method for Data Selection. By systematically modeling error patterns through a score transition matrix, DS2 corrects LLM-based scores and promotes diversity in the selected data samples. Our approach shows that a curated subset (just 3.3% of the original dataset) outperforms full-scale datasets (300k samples) across various machine-alignment benchmarks, and matches or surpasses human-aligned datasets such as LIMA with the same sample size (1k samples). These findings challenge conventional data scaling assumptions, highlighting that redundant, low-quality samples can degrade performance and reaffirming that "more can be less."

With the rapid advancement of large language models (LLMs), there is a pressing need for a comprehensive evaluation suite to assess their capabilities and limitations. Existing LLM leaderboards often reference scores reported in other papers without consistent settings and prompts, which may inadvertently encourage cherry-picking favored settings and prompts for better results. In this work, we introduce GPT-Fathom, an open-source and reproducible LLM evaluation suite built on top of OpenAI Evals. We systematically evaluate 10+ leading LLMs as well as OpenAI's legacy models on 20+ curated benchmarks across 7 capability categories, all under aligned settings. Our retrospective study on OpenAI's earlier models offers valuable insights into the evolutionary path from GPT-3 to GPT-4. Currently, the community is eager to know how GPT-3 progressively improves to GPT-4, including technical details like whether adding code data improves LLM's reasoning capability, which aspects of LLM capability can be improved by SFT and RLHF, how much is the alignment tax, etc. Our analysis sheds light on many of these questions, aiming to improve the transparency of advanced LLMs.

How can we use AI to discover a new state of the art for a scientific problem? Prior work in test-time scaling, such as AlphaEvolve, performs search by prompting a frozen LLM. We perform reinforcement learning at test time, so the LLM can continue to train, but now with experience specific to the test problem. This form of continual learning is quite special, because its goal is to produce one great solution rather than many good ones on average, and to solve this very problem rather than generalize to other problems. Therefore, our learning objective and search subroutine are designed to prioritize the most promising solutions. We call this method Test-Time Training to Discover (TTT-Discover). Following prior work, we focus on problems with continuous rewards. We report results for every problem we attempted, across mathematics, GPU kernel engineering, algorithm design, and biology. TTT-Discover sets the new state of the art in almost all of them: (i) Erdős' minimum overlap problem and an autocorrelation inequality; (ii) a GPUMode kernel competition (up to 2times faster than prior art); (iii) past AtCoder algorithm competitions; and (iv) denoising problem in single-cell analysis. Our solutions are reviewed by experts or the organizers. All our results are achieved with an open model, OpenAI gpt-oss-120b, and can be reproduced with our publicly available code, in contrast to previous best results that required closed frontier models. Our test-time training runs are performed using Tinker, an API by Thinking Machines, with a cost of only a few hundred dollars per problem.

Some security-development settings require local models that map an objective to an ordered, checkable plan. We present a low-rank decomposition adapter and release a case-disjoint corpus with 24 authored decompositions and 83 derived next-step examples across 24 satellite-security cases. To prevent reference-plan leakage, prompts contain one objective and eight shuffled SPARTA candidates. The evaluation candidate sets are constructed with oracle inclusion of every reference technique, so this study measures selection from a controlled set, not retrieval. On six fixed cases and five decoding seeds, mean scores vary across adapters and prompted Qwen2.5 baselines; a separate greedy check over three adapter training seeds shows non-negligible training variance. A 24-fold greedy leave-one-case-out control at the two compact sizes likewise shows tradeoffs: at 1.5B, adapter precision is 0.583 against 0.480 for two-shot prompting, while recall is 0.586 against 0.660. Case-resampling intervals for adapter precision relative to the strongest prompted baseline cross zero at 1.5B and 7B. The adapter is also substantially more format-compliant, which prevents attributing all score differences to selection. Autoregressive next-step accuracy is 0.08, 0.06, and 0.29 at 0.5B, 1.5B, and 7B. We therefore offer the formulation, leakage-controlled dataset, reproducibility package, and descriptive analysis as a proof of concept, not evidence of a reliable stand-alone decomposition system or a general adapter advantage. A separate diagnostic tests contract reuse.

Bug reports often lack sufficient detail for developers to reproduce and fix the underlying defects. Bug Reproduction Tests (BRTs), tests that fail when the bug is present and pass when it has been resolved, are crucial for debugging, but they are rarely included in bug reports, both in open-source and in industrial settings. Thus, automatically generating BRTs from bug reports has the potential to accelerate the debugging process and lower time to repair. This paper investigates automated BRT generation within an industry setting, specifically at Google, focusing on the challenges of a large-scale, proprietary codebase and considering real-world industry bugs extracted from Google's internal issue tracker. We adapt and evaluate a state-of-the-art BRT generation technique, LIBRO, and present our agent-based approach, BRT Agent, which makes use of a fine-tuned Large Language Model (LLM) for code editing. Our BRT Agent significantly outperforms LIBRO, achieving a 28% plausible BRT generation rate, compared to 10% by LIBRO, on 80 human-reported bugs from Google's internal issue tracker. We further investigate the practical value of generated BRTs by integrating them with an Automated Program Repair (APR) system at Google. Our results show that providing BRTs to the APR system results in 30% more bugs with plausible fixes. Additionally, we introduce Ensemble Pass Rate (EPR), a metric which leverages the generated BRTs to select the most promising fixes from all fixes generated by APR system. Our evaluation on EPR for Top-K and threshold-based fix selections demonstrates promising results and trade-offs. For example, EPR correctly selects a plausible fix from a pool of 20 candidates in 70% of cases, based on its top-1 ranking.

Tumor protein P53 is believed to be involved in over half of human cancers cases, the prediction of malignancies plays essential roles not only in advance detection for cancer, but also in discovering effective prevention and treatment of cancer, till now there isn't approach be able in prediction the mutated in tumor protein P53 which is caused high ratio of human cancers like breast, Blood, skin, liver, lung, bladder etc. This research proposed a new approach for prediction pre-cancer via detection malignant mutations in tumor protein P53 using bioinformatics tools like FASTA, BLAST, CLUSTALW and TP53 databases worldwide. Implement and apply this new approach of prediction pre-cancer through mutations at tumor protein P53 shows an effective result when used more specific parameters/features to extract the prediction result that means when the user increase the number of filters of the results which obtained from the database gives more specific diagnosis and classify, addition that the detecting pre-cancer via prediction mutated tumor protein P53 will reduces a person's cancers in the future by avoiding exposure to toxins, radiation or monitoring themselves at older ages by change their food, environment, even the pace of living. Also that new approach of prediction pre-cancer will help if there is any treatment can give for that person to therapy the mutated tumor protein P53. Index Terms (Normal Homology TP53 gene, Tumor Protein P53, Oncogene Labs, GC and AT content, FASTA, BLAST, ClustalW)

Large language models (LLMs) hold great promise in summarizing medical evidence. Most recent studies focus on the application of proprietary LLMs. Using proprietary LLMs introduces multiple risk factors, including a lack of transparency and vendor dependency. While open-source LLMs allow better transparency and customization, their performance falls short compared to proprietary ones. In this study, we investigated to what extent fine-tuning open-source LLMs can further improve their performance in summarizing medical evidence. Utilizing a benchmark dataset, MedReview, consisting of 8,161 pairs of systematic reviews and summaries, we fine-tuned three broadly-used, open-sourced LLMs, namely PRIMERA, LongT5, and Llama-2. Overall, the fine-tuned LLMs obtained an increase of 9.89 in ROUGE-L (95% confidence interval: 8.94-10.81), 13.21 in METEOR score (95% confidence interval: 12.05-14.37), and 15.82 in CHRF score (95% confidence interval: 13.89-16.44). The performance of fine-tuned LongT5 is close to GPT-3.5 with zero-shot settings. Furthermore, smaller fine-tuned models sometimes even demonstrated superior performance compared to larger zero-shot models. The above trends of improvement were also manifested in both human and GPT4-simulated evaluations. Our results can be applied to guide model selection for tasks demanding particular domain knowledge, such as medical evidence summarization.

Distinct-n scoreLi2016 is a widely used automatic metric for evaluating diversity in language generation tasks. However, we observed that the original approach for calculating distinct scores has evident biases that tend to assign higher penalties to longer sequences. We refine the calculation of distinct scores by scaling the number of distinct tokens based on their expectations. We provide both empirical and theoretical evidence to show that our method effectively removes the biases existing in the original distinct score. Our experiments show that our proposed metric, Expectation-Adjusted Distinct (EAD), correlates better with human judgment in evaluating response diversity. To foster future research, we provide an example implementation at https://github.com/lsy641/Expectation-Adjusted-Distinct.

Antibody lead optimization calls for a small, bounded set of edits to an existing candidate: substitutions, but also insertions and deletions. Edit-based generative models are the only ones that allocate such an edit budget without fixing the edit positions, the edit count, or the output length in advance. However, the existing approaches Edit Flows and EvoFlows did not release code or complete training specifications. Here, we show that both methods follow the same underlying process -- edits firing one at a time, at learned rates, in continuous time -- the pure-jump case of generator matching over finite sequences. With EditJumps we introduce the first open implementation of this framework, with a single generalist antibody editor trained on 1.66M Observed Antibody Space homolog pairs to propose homolog-like variants of a seed sequence, editing unseen leads zero-shot, without the per-family retraining original approaches require. Replicating this system from scratch exposes why open code is essential for generative biology: reconciling published edit distributions required reverse-engineering an undocumented rate-scaling hyperparameter that dictates realized mutation counts. Moreover, we show that published evaluation metrics are highly sensitive to reference sample size, frequently flipping method rankings. We release our full codebase, automated test suite, and configurations at: https://github.com/VisiumCH/editjumps

In recent years, cancer genome sequencing and other high-throughput studies of cancer genomes have generated many notable discoveries. In this review, Novel genomic alteration mechanisms, such as chromothripsis (chromosomal crisis) and kataegis (mutation storms), and their implications for cancer are discussed. Genomic alterations spur cancer genome evolution. Thus, the relationship between cancer clonal evolution and cancer stems cells is commented. The key question in cancer biology concerns how these genomic alterations support cancer development and metastasis in the context of biological functioning. Thus far, efforts such as pathway analysis have improved the understanding of the functional contributions of genetic mutations and DNA copy number variations to cancer development, progression and metastasis. However, the known pathways correspond to a small fraction, plausibly 5-10%, of somatic mutations and genes with an altered copy number. To develop a comprehensive understanding of the function of these genomic alterations in cancer, an integrative network framework is proposed and discussed. Finally, the challenges and the directions of studying cancer omic data using an integrative network approach are commented.

There are various benchmarks to evaluate bugfixing capabilities of Large Language Models. However, most widespread benchmarks do not fully reflect real-world bugfixing practices. They are small, weakening statistical reliability, and the buggy programs are often similar to one another, potentially distorting evaluation results. The range of bug types can also be narrow, failing to capture a representative range of bugs. To address these issues, we introduce MegaBugFix, a large-scale bugfixing benchmark containing 12,629 buggy Python programs synthesized from correct ones by a Large Language Model. Bug injections were generated as diffs representing code changes. Through this approach, we were able to avoid common pitfalls of LLM-based mutation techniques like injecting overly simplistic bugs or failing to modify the input program. We evaluated 13 open-weight models on MegaBugFix and baseline benchmarks, finding consistently lower performance on MegaBugFix. This reveals that our benchmark presents more challenging bugs and exposes model failures that may remain hidden when evaluating on existing benchmarks. The benchmark and fine-tuned model used for bug injection are available at hf.co/collections/szalontaib/megabugfix.

Recent advances in automated test generation utilises language models to produce unit tests. While effective, language models tend to generate many incorrect tests with respect to both syntax and semantics. Although such incorrect tests can be easily detected and discarded, they constitute a "missed opportunity" -- if fixed, they are often valuable as they directly add testing value (they effectively target the underlying program logic to be tested) and indirectly form good seeds for generating additional tests. To this end, we propose a simple technique for repairing some of these incorrect tests through a combination of rule-based static analysis and re-prompting. We evaluate this simple approach, named YATE, on a set of 6 open-source projects and show that it can effectively produce tests that cover on average 32.06% more lines and kill 21.77% more mutants than a plain LLM-based method. We also compare YATE with four other LLM-based methods, namely HITS, SYMPROMPT, TESTSPARK and COVERUP and show that it produces tests that cover substantially more code. YATE achieves 22% higher line coverage, 20% higher branch coverage and kill 20% more mutants at a comparable cost (number of calls to LLMs).

Scientific discovery is an inherently creative and uncertain process, requiring reasoning beyond the recall of known knowledge. While many benchmarks have been proposed to evaluate large language model (LLM) performance on deep research tasks via multi-hop retrieval, their innovative reasoning abilities essential for true scientific discovery remain largely untested. We introduce a benchmark framework for evaluating model performance in scientific discovery and reasoning, building up from a raw problem to the classical null hypothesis test. In our framework, models initially receive only the topic and research question from a recent paper, with technical details progressively revealed. At each stage of information disclosure, the model is tasked with generating hypotheses that address the research question, which is compared with the conclusions from the original paper and evaluated via automated semantic similarity of constituent atomic claims. This progressive evaluation of semantic divergence from ground-truth conclusions enables assessment of a model's innovativeness (under minimal information) to grounded reasoning capabilities (under full experimental details), both critical for using LLMs for scientific discovery purposes. Our framework provides a foundation for systematically evaluating scientific reasoning and discovery capabilities in LLMs, crucial for advancing the development of next-generation AI scientist/co-scientist systems. Specifically, here we evaluate GPT-5, GPT-5.4, Gemini 2.5 pro, and Gemini 3.1 pro preview across 45 papers spanning bioactive materials, mechanical materials, and nanomaterials. We find that GPT-5.4 and Gemini 3.1 pro outperform their previous generation counterparts as expected, and GPT-5.4 in particular maintains 0.7 F1 score alignment with ground truth conclusions even under minimal context.

We introduce DoGBENCH (Documentation Generation Benchmark), to our knowledge, the first benchmark for generating and maintaining real user-facing software documentation. It asks whether an agent can produce documentation that experienced technical writers would accept in review. The benchmark contains 292 items from open source projects, including Helm, PostHog, and Mautic. Each item gives the agent a pre-change repository and a trigger, such as a code pull request or a reported documentation gap. The agent must first decide whether the documentation needs an update. For items that need one, the agent must produce an acceptable patch in one attempt. For items that do not need updates, the agent must abstain. Task-specific rubrics, validated with project maintainers, score each patch on accuracy, completeness, reader guidance, placement, and repository conventions. The composite score combines patch quality with correct abstention, and a score of 100 means an agent meets every requirement for the task. Scores should not be interpreted as a percentage of an expert's capability. We evaluated seven agents. The highest-scoring agent reached 47.3 out of 100 on the 117-item held-out split. In a separate audit of 1,267 patches, the most common failure modes were task-completion gaps (45.5%), technical inaccuracies (36.6%), and incomplete conceptual or reference coverage (32.5%). Analysis of the corresponding trajectories identified three key patterns associated with these failures: (1) describing interfaces without examining how readers use them (36.0%), (2) missing decisive evidence and filling the gaps with plausible assumptions (33.1%), and (3) stopping after finding the first plausible documentation surface and leaving other affected pages stale (30.1%).

Large Language Models (LLMs) have shown promise in accelerating the scientific research pipeline. A key capability for this process is the ability to generate novel research ideas, and prior studies have found settings in which LLM-generated research ideas were judged as more novel than human-expert ideas. However, a good idea should not simply appear to be novel, it should also result in better research after being executed. To test whether AI-generated ideas lead to better research outcomes, we conduct an execution study by recruiting 43 expert researchers to execute randomly-assigned ideas, either written by experts or generated by an LLM. Each expert spent over 100 hours implementing the idea and wrote a 4-page short paper to document the experiments. All the executed projects are then reviewed blindly by expert NLP researchers. Comparing the review scores of the same ideas before and after execution, the scores of the LLM-generated ideas decrease significantly more than expert-written ideas on all evaluation metrics (novelty, excitement, effectiveness, and overall; p < 0.05), closing the gap between LLM and human ideas observed at the ideation stage. When comparing the aggregated review scores from the execution study, we even observe that for many metrics there is a flip in rankings where human ideas score higher than LLM ideas. This ideation-execution gap highlights the limitations of current LLMs in generating truly effective research ideas and the challenge of evaluating research ideas in the absence of execution outcomes.

Code Large Language Models (CodeLLMs) have ushered in a new era of code generation advancements. However, selecting the best solutions from among all possible CodeLLM solutions remains a challenge. Previous methods frequently overlooked the intricate functional similarities and interactions between clusters, resulting in suboptimal results. In this work, we introduce SRank, a novel reranking strategy for selecting the best solution from code generation that focuses on modeling inter-cluster relationship. By quantifying the functional overlap between clusters, our approach provides a better ranking strategy of code solutions. Empirical results show that our method achieves a remarkable results on pass@1 score. For instance, on the Human-Eval benchmark, we achieve 69.66\% in pass@1 with Codex002, 75.31\% for WizardCoder, 53.99\% for StarCoder and 60.55\% for CodeGen, which surpass the state-of-the-arts solution ranking methods, such as CodeT and Coder-Reviewer on the same CodeLLM with significant margin (approx 6.1% improvement on average). Comparing to the random sampling method, we can achieve an average improvement of approx 23.07% on Human-Eval and 17.64\% on MBPP. Even in scenarios with limited test inputs, our approach demonstrates robustness and superiority, marking a new state-of-the-arts in code generation reranking.

We introduce UnpredictaBench, an evaluation that tests the ability of large language models (LLMs) to capture true underlying distributions. As LLMs are increasingly used as substitutes for other entities (e.g., for humans in economic simulations), the tendency of many models to collapse towards a single plausible answer means a failure to capture the unpredictability of real systems. Recent work on improving output diversity is insufficient for this setting: simulation requires samples that are calibrated to a target distribution, not merely varied outputs. UnpredictaBench isolates a simplified but fundamental version of this problem: sampling outcomes from individual target distributions, including canonical statistical distributions, distributions induced by stochastic programs, and natural-language scenarios that describe random processes. We introduce 448 such problems together with KS@N, a general-purpose evaluation metric that quantifies how well a model outputs approximate black-box target distributions via the Kolmogorov-Smirnov statistical test. This is the rate at which we fail to reject model samples of size N against ground-truth samples, with larger N indicating greater difficulty. Tested across open and proprietary models, we find a large spread in distributional capabilities. For instance, when models generate samples of size 100 (KS@100, our standard metric), scores range from near 0 to over 20%. No model is able to achieve over 40% at KS@100, showing significant headroom in distributional sampling as a capability. Although adding reasoning can somewhat increase scores, we find no immediate solution for this issue. UnpredictaBench shows that even simple distributional simulation remains challenging, making it a necessary first step toward using LLMs as stand-ins for complex systems.

Peripheral blood smears remain a cornerstone in the diagnosis of hematological neoplasms, offering rapid and valuable insights that inform subsequent diagnostic steps. However, since neoplastic transformations typically arise in the bone marrow, they may not manifest as detectable aberrations in peripheral blood, presenting a diagnostic challenge. In this paper, we introduce cAItomorph, an explainable transformer-based AI model, trained to classify hematological malignancies based on peripheral blood cytomorphology. Our data comprises peripheral blood single-cell images from 6115 patients with diagnoses confirmed by cytomorphology, cytogenetics, molecular genetics, and immunophenotyping from bone marrow samples, and 495 healthy controls, eight coarse classes. cAItomorph leverages the DinoBloom hematology foundation model and aggregates image encodings via a transformer-based architecture into a single vector. It achieves an overall accuracy of 0.72 in eight disease classification, with F1 scores of 0.76 for acute leukemia, 0.80 for myeloproliferative neoplasms and 0.94 for healthy cases. The overall accuracy increases to 0.87 in top-2 predictions. cAItomorph achieves high sensitivity for acute leukemia cases in external test sets. By analyzing attention heads, we demonstrate clinically relevant cell-level attentions in both internal and external test sets. Moreover, our model's calibrated prediction probabilities reduce the false discovery rate from 13.5% to 8.7% without missing any acute leukemia cases, thereby decreasing the number of unnecessary bone marrow aspirations based on peripheral blood smears. This study highlights the potential of AI-assisted diagnostics in hematological malignancies, illustrating how models trained on real-world data could enhance diagnostic accuracy and reduce invasive procedures.

Leaderboards rank long-horizon agents by their final outputs. Yet a higher score alone does not establish whether two systems are comparable or which stage accounts for the difference. Unequal evidence, inputs, or budgets can affect scores, and statistical corrections do not remove this mismatch. We introduce compositional controllability to address these questions. A comparison window covers one stage, several stages, or the whole agent. Our central result bounds the gap between observed and controlled score differences using only nuisance outside the window. This yields an admissibility test applied before scores are inspected. Inadmissible comparisons are refused. For admissible pairs, an ordering is certified only when the score gap exceeds the combined sampling and nuisance radii; otherwise, it remains undecided. These decisions give each system a rank interval. We introduce BioLitBench, a benchmark of 2,042 biomedical articles represented as structured claim graphs. Among seven published pipelines, a conventional statistical analysis declares a winner in 14 of 21 pairwise comparisons. Yet the top-ranked system alone received the target review's bibliography. To isolate pipeline performance, our test requires matched inputs and a fixed backbone model. It refuses 11 of the 21 comparisons, including every comparison involving the top-ranked system. Seven of the 14 conventional conclusions fall within these refused pairs. The same comparison windows support stage-level training. We train SCRIBE on Qwen3.8-27B using rewards measured at each stage's exit. Under matched evidence, SCRIBE achieves a certified rank interval of [1,2], with certified advantages over all evaluated published pipelines and the evaluated Claude and OpenAI agents. Under same pool, SCRIBE matches the strongest published retriever and is certified above three published pipelines.