arXiv subjects

Martin Josifoski

Publications and source records attributed to Martin Josifoski.

At least 19 recordsLinked to original sources

arXiv2026

AI Research Preference Models

AI research agents (AIRA) can now carry machine learning experiments from proposal through implementation and evaluation. Yet progress on frontier tasks is throttled by the cost of evaluations that can consume days of GPU time. When an agent can propose far more candidates than it can afford to run, progress depends on its research preference: how it allocates a fixed execution budget across many candidates. We introduce AI Research Preference Models (RPMs) that predict which candidate solution is most promising, without paying the cost of running them all. We build RPMs from frozen pretrained language models in two variants: an inference-only model that reasons over candidate plans, code, and previously executed solutions, and an agentic model that additionally runs small-scale pilot experiments. Integrated into the AIRA-dojo research agent and evaluated on the machine learning research benchmark AIRS-Bench, the two variants increase the average normalized score from 0.684 to 0.711 and 0.729, respectively. Both reach the unguided agent's 24-hour performance in roughly 15 hours, using less than two-thirds of its execution budget, and together yield new state-of-the-art results on two AIRS-Bench tasks.

cs.AI↗

arXiv2026

AIRA_2: Overcoming Bottlenecks in AI Research Agents

Existing research has identified three structural performance bottlenecks in AI research agents: (1) synchronous single-GPU execution constrains sample throughput, limiting the benefit of search; (2) a generalization gap where validation-based selection causes overfitting and performance to degrade over extended search horizons; and (3) the limited capability of fixed, single-turn LLM operators imposes a ceiling on search performance. We introduce AIRA$_2$, which addresses these bottlenecks through three architectural choices: an asynchronous multi-GPU worker pool that increases experiment throughput linearly; a Hidden Consistent Evaluation protocol that delivers a reliable evaluation signal; and ReAct agents that dynamically scope their actions and debug interactively. On MLE-bench-30, AIRA$^{\dagger}_{2}$ achieves a mean Percentile Rank of 81.5% at 24 hours and 83.1% at 72 hours, outperforming the strongest baseline, which achieves 72.7%. On AIRS-Bench, AIRA$_2$ exceeds human state-of-the-art on 6 out of 20 diverse research tasks. Ablations confirm that each architectural component is necessary, that performance follows a predictable scaling law that transfers across LLM backbones, and that the "overfitting" reported in prior work was driven by evaluation noise rather than true data memorization.

cs.AI↗

arXiv2026

Evaluating Language Model Agency through Negotiations

We introduce an approach to evaluate language model (LM) agency using negotiation games. This approach better reflects real-world use cases and addresses some of the shortcomings of alternative LM benchmarks. Negotiation games enable us to study multi-turn, and cross-model interactions, modulate complexity, and side-step accidental evaluation data leakage. We use our approach to test six widely used and publicly accessible LMs, evaluating performance and alignment in both self-play and cross-play settings. Noteworthy findings include: (i) only closed-source models tested here were able to complete these tasks; (ii) cooperative bargaining games proved to be most challenging to the models; and (iii) even the most powerful models sometimes "lose" to weaker opponents

cs.CL↗

arXiv2026

AIRS-Bench: a Suite of Tasks for Frontier AI Research Science Agents

LLM agents hold significant promise for advancing scientific research. To accelerate this progress, we introduce AIRS-Bench (the AI Research Science Benchmark), a suite of 20 tasks sourced from state-of-the-art machine learning papers. These tasks span diverse domains, including language modeling, mathematics, bioinformatics, and time series forecasting. AIRS-Bench tasks assess agentic capabilities over the full research lifecycle -- including idea generation, experiment analysis and iterative refinement -- without providing baseline code. The AIRS-Bench task format is versatile, enabling easy integration of new tasks and rigorous comparison across different agentic frameworks. We establish baselines using frontier models paired with both sequential and parallel scaffolds. Our results show that agents exceed human SOTA in four tasks but fail to match it in sixteen others. Even when agents surpass human benchmarks, they do not reach the theoretical performance ceiling for the underlying tasks. These findings indicate that AIRS-Bench is far from saturated and offers substantial room for improvement. We open-source the AIRS-Bench task definitions and evaluation code to catalyze further development in autonomous scientific research.

cs.AI↗

arXiv2025

What Does It Take to Be a Good AI Research Agent? Studying the Role of Ideation Diversity

AI research agents offer the promise to accelerate scientific progress by automating the design, implementation, and training of machine learning models. However, the field is still in its infancy, and the key factors driving the success or failure of agent trajectories are not fully understood. We examine the role that ideation diversity plays in agent performance. First, we analyse agent trajectories on MLE-bench, a well-known benchmark to evaluate AI research agents, across different models and agent scaffolds. Our analysis reveals that different models and agent scaffolds yield varying degrees of ideation diversity, and that higher-performing agents tend to have increased ideation diversity. Further, we run a controlled experiment where we modify the degree of ideation diversity, demonstrating that higher ideation diversity results in stronger performance. Finally, we strengthen our results by examining additional evaluation metrics beyond the standard medal-based scoring of MLE-bench, showing that our findings still hold across other agent performance metrics.

cs.AI↗

arXiv2025

AI Research Agents for Machine Learning: Search, Exploration, and Generalization in MLE-bench

AI research agents are demonstrating great potential to accelerate scientific progress by automating the design, implementation, and training of machine learning models. We focus on methods for improving agents' performance on MLE-bench, a challenging benchmark where agents compete in Kaggle competitions to solve real-world machine learning problems. We formalize AI research agents as search policies that navigate a space of candidate solutions, iteratively modifying them using operators. By designing and systematically varying different operator sets and search policies (Greedy, MCTS, Evolutionary), we show that their interplay is critical for achieving high performance. Our best pairing of search strategy and operator set achieves a state-of-the-art result on MLE-bench lite, increasing the success rate of achieving a Kaggle medal from 39.6% to 47.7%. Our investigation underscores the importance of jointly considering the search strategy, operator design, and evaluation methodology in advancing automated machine learning.

cs.AI↗

arXiv2025

Interactive Evaluation of Large Language Models for Multi-Requirement Software Engineering Tasks

Standard single-turn, static benchmarks fall short in evaluating the nuanced capabilities of Large Language Models (LLMs) on complex tasks such as software engineering. In this work, we propose a novel interactive evaluation framework that assesses LLMs on multi-requirement programming tasks through structured, feedback-driven dialogue. Each task is modeled as a requirement dependency graph, and an ``interviewer'' LLM, aware of the ground-truth solution, provides minimal, targeted hints to an ``interviewee'' model to help correct errors and fulfill target constraints. This dynamic protocol enables fine-grained diagnostic insights into model behavior, uncovering strengths and systematic weaknesses that static benchmarks fail to measure. We build on DevAI, a benchmark of 55 curated programming tasks, by adding ground-truth solutions and evaluating the relevance and utility of interviewer hints through expert annotation. Our results highlight the importance of dynamic evaluation in advancing the development of collaborative code-generating agents.

cs.AI↗

arXiv2025

The Automated LLM Speedrunning Benchmark: Reproducing NanoGPT Improvements

Rapid advancements in large language models (LLMs) have the potential to assist in scientific progress. A critical capability toward this endeavor is the ability to reproduce existing work. To evaluate the ability of AI agents to reproduce results in an active research area, we introduce the Automated LLM Speedrunning Benchmark, leveraging the research community contributions on the NanoGPT speedrun, a competition to train a GPT-2 model in the shortest time. Each of the 19 speedrun tasks provides the agent with the previous records training script, optionally paired with one of three hint formats, ranging from pseudocode to paper-like descriptions of the new records improvements. Records execute quickly by design and speedrun improvements encompass diverse code-level changes, ranging from high-level algorithmic advancements to hardware-aware optimizations. These features make the benchmark both accessible and realistic for the frontier problem of improving LLM training. We find that recent reasoning LLMs combined with SoTA scaffolds struggle to reimplement already-known innovations in our benchmark, even when given detailed hints. Our benchmark thus provides a simple, non-saturated measure of an LLMs ability to automate scientific reproduction, a necessary (but not sufficient) skill for an autonomous research agent.

cs.AI↗

arXiv2025

Agentic AI: The Era of Semantic Decoding

Recent work demonstrated great promise in the idea of orchestrating collaborations between LLMs, human input, and various tools to address the inherent limitations of LLMs. We propose a novel perspective called semantic decoding, which frames these collaborative processes as optimization procedures in semantic space. Specifically, we conceptualize LLMs as semantic processors that manipulate meaningful pieces of information that we call semantic tokens (known thoughts). LLMs are among a large pool of other semantic processors, including humans and tools, such as search engines or code executors. Collectively, semantic processors engage in dynamic exchanges of semantic tokens to progressively construct high-utility outputs. We refer to these orchestrated interactions among semantic processors, optimizing and searching in semantic space, as semantic decoding algorithms. This concept draws a direct parallel to the well-studied problem of syntactic decoding, which involves crafting algorithms to best exploit auto-regressive language models for extracting high-utility sequences of syntactic tokens. By focusing on the semantic level and disregarding syntactic details, we gain a fresh perspective on the engineering of AI systems, enabling us to imagine systems with much greater complexity and capabilities. In this position paper, we formalize the transition from syntactic to semantic tokens as well as the analogy between syntactic and semantic decoding. Subsequently, we explore the possibilities of optimizing within the space of semantic tokens via semantic decoding algorithms. We conclude with a list of research opportunities and questions arising from this fresh perspective. The semantic decoding perspective offers a powerful abstraction for search and optimization directly in the space of meaningful concepts, with semantic tokens as the fundamental units of a new type of computation.

cs.CL↗

arXiv2024

Sketch-Guided Constrained Decoding for Boosting Blackbox Large Language Models without Logit Access

Constrained decoding, a technique for enforcing constraints on language model outputs, offers a way to control text generation without retraining or architectural modifications. Its application is, however, typically restricted to models that give users access to next-token distributions (usually via softmax logits), which poses a limitation with blackbox large language models (LLMs). This paper introduces sketch-guided constrained decoding (SGCD), a novel approach to constrained decoding for blackbox LLMs, which operates without access to the logits of the blackbox LLM. SGCD utilizes a locally hosted auxiliary model to refine the output of an unconstrained blackbox LLM, effectively treating this initial output as a "sketch" for further elaboration. This approach is complementary to traditional logit-based techniques and enables the application of constrained decoding in settings where full model transparency is unavailable. We demonstrate the efficacy of SGCD through experiments in closed information extraction and constituency parsing, showing how it enhances the utility and flexibility of blackbox LLMs for complex NLP tasks.

cs.CL↗

arXiv2024

A Glitch in the Matrix? Locating and Detecting Language Model Grounding with Fakepedia

Large language models (LLMs) have an impressive ability to draw on novel information supplied in their context. Yet the mechanisms underlying this contextual grounding remain unknown, especially in situations where contextual information contradicts factual knowledge stored in the parameters, which LLMs also excel at recalling. Favoring the contextual information is critical for retrieval-augmented generation methods, which enrich the context with up-to-date information, hoping that grounding can rectify outdated or noisy stored knowledge. We present a novel method to study grounding abilities using Fakepedia, a novel dataset of counterfactual texts constructed to clash with a model's internal parametric knowledge. In this study, we introduce Fakepedia, a counterfactual dataset designed to evaluate grounding abilities when the internal parametric knowledge clashes with the contextual information. We benchmark various LLMs with Fakepedia and conduct a causal mediation analysis of LLM components when answering Fakepedia queries, based on our Masked Grouped Causal Tracing (MGCT) method. Through this analysis, we identify distinct computational patterns between grounded and ungrounded responses. We finally demonstrate that distinguishing grounded from ungrounded responses is achievable through computational analysis alone. Our results, together with existing findings about factual recall mechanisms, provide a coherent narrative of how grounding and factual recall mechanisms interact within LLMs.

cs.CL↗

arXiv2024

Symbolic Autoencoding for Self-Supervised Sequence Learning

Traditional language models, adept at next-token prediction in text sequences, often struggle with transduction tasks between distinct symbolic systems, particularly when parallel data is scarce. Addressing this issue, we introduce \textit{symbolic autoencoding} ($Σ$AE), a self-supervised framework that harnesses the power of abundant unparallel data alongside limited parallel data. $Σ$AE connects two generative models via a discrete bottleneck layer and is optimized end-to-end by minimizing reconstruction loss (simultaneously with supervised loss for the parallel data), such that the sequence generated by the discrete bottleneck can be read out as the transduced input sequence. We also develop gradient-based methods allowing for efficient self-supervised sequence learning despite the discreteness of the bottleneck. Our results demonstrate that $Σ$AE significantly enhances performance on transduction tasks, even with minimal parallel data, offering a promising solution for weakly supervised learning scenarios.

cs.LG↗

arXiv2024

Flows: Building Blocks of Reasoning and Collaborating AI

Recent advances in artificial intelligence (AI) have produced highly capable and controllable systems. This creates unprecedented opportunities for structured reasoning as well as collaboration among multiple AI systems and humans. To fully realize this potential, it is essential to develop a principled way of designing and studying such structured interactions. For this purpose, we introduce the conceptual framework Flows. Flows are self-contained building blocks of computation, with an isolated state, communicating through a standardized message-based interface. This modular design simplifies the process of creating Flows by allowing them to be recursively composed into arbitrarily nested interactions and is inherently concurrency-friendly. Crucially, any interaction can be implemented using this framework, including prior work on AI-AI and human-AI interactions, prompt engineering schemes, and tool augmentation. We demonstrate the potential of Flows on competitive coding, a challenging task on which even GPT-4 struggles. Our results suggest that structured reasoning and collaboration substantially improve generalization, with AI-only Flows adding +21 and human-AI Flows adding +54 absolute points in terms of solve rate. To support rapid and rigorous research, we introduce the aiFlows library embodying Flows. The aiFlows library is available at https://github.com/epfl-dlab/aiflows. Data and Flows for reproducing our experiments are available at https://github.com/epfl-dlab/cc_flows.

cs.AI↗

arXiv2024

Grammar-Constrained Decoding for Structured NLP Tasks without Finetuning

Despite their impressive performance, large language models (LMs) still struggle with reliably generating complex output structures when not finetuned to follow the required output format exactly. To address this issue, grammar-constrained decoding (GCD) can be used to control the generation of LMs, guaranteeing that the output follows a given structure. Most existing GCD methods are, however, limited to specific tasks, such as parsing or code generation. In this work, we demonstrate that formal grammars can describe the output space for a much wider range of tasks and argue that GCD can serve as a unified framework for structured NLP tasks in general. For increased flexibility, we introduce input-dependent grammars, which allow the grammar to depend on the input and thus enable the generation of different output structures for different inputs. We then empirically demonstrate the power and flexibility of GCD-enhanced LMs on (1) information extraction, (2) entity disambiguation, and (3) constituency parsing. Our results indicate that grammar-constrained LMs substantially outperform unconstrained LMs or even beat task-specific finetuned models. Grammar constraints thus hold great promise for harnessing off-the-shelf LMs for a wide range of structured NLP tasks, especially where training data is scarce or finetuning is expensive. Code and data: https://github.com/epfl-dlab/GCD.

cs.CL↗

arXiv2023

Scalable PAC-Bayesian Meta-Learning via the PAC-Optimal Hyper-Posterior: From Theory to Practice

Meta-Learning aims to speed up the learning process on new tasks by acquiring useful inductive biases from datasets of related learning tasks. While, in practice, the number of related tasks available is often small, most of the existing approaches assume an abundance of tasks; making them unrealistic and prone to overfitting. A central question in the meta-learning literature is how to regularize to ensure generalization to unseen tasks. In this work, we provide a theoretical analysis using the PAC-Bayesian theory and present a generalization bound for meta-learning, which was first derived by Rothfuss et al. (2021a). Crucially, the bound allows us to derive the closed form of the optimal hyper-posterior, referred to as PACOH, which leads to the best performance guarantees. We provide a theoretical analysis and empirical case study under which conditions and to what extent these guarantees for meta-learning improve upon PAC-Bayesian per-task learning bounds. The closed-form PACOH inspires a practical meta-learning approach that avoids the reliance on bi-level optimization, giving rise to a stochastic optimization problem that is amenable to standard variational methods that scale well. Our experiments show that, when instantiating the PACOH with Gaussian processes and Bayesian Neural Networks models, the resulting methods are more scalable, and yield state-of-the-art performance, both in terms of predictive accuracy and the quality of uncertainty estimates.

stat.ML↗

arXiv2023

Exploiting Asymmetry for Synthetic Training Data Generation: SynthIE and the Case of Information Extraction

Large language models (LLMs) have great potential for synthetic data generation. This work shows that useful data can be synthetically generated even for tasks that cannot be solved directly by LLMs: for problems with structured outputs, it is possible to prompt an LLM to perform the task in the reverse direction, by generating plausible input text for a target output structure. Leveraging this asymmetry in task difficulty makes it possible to produce large-scale, high-quality data for complex tasks. We demonstrate the effectiveness of this approach on closed information extraction, where collecting ground-truth data is challenging, and no satisfactory dataset exists to date. We synthetically generate a dataset of 1.8M data points, establish its superior quality compared to existing datasets in a human evaluation, and use it to finetune small models (220M and 770M parameters), termed SynthIE, that outperform the prior state of the art (with equal model size) by a substantial margin of 57 absolute points in micro-F1 and 79 points in macro-F1. Code, data, and models are available at https://github.com/epfl-dlab/SynthIE.

cs.CL↗

arXiv2023

Generating Faithful Synthetic Data with Large Language Models: A Case Study in Computational Social Science

Large Language Models (LLMs) have democratized synthetic data generation, which in turn has the potential to simplify and broaden a wide gamut of NLP tasks. Here, we tackle a pervasive problem in synthetic data generation: its generative distribution often differs from the distribution of real-world data researchers care about (in other words, it is unfaithful). In a case study on sarcasm detection, we study three strategies to increase the faithfulness of synthetic data: grounding, filtering, and taxonomy-based generation. We evaluate these strategies using the performance of classifiers trained with generated synthetic data on real-world data. While all three strategies improve the performance of classifiers, we find that grounding works best for the task at hand. As synthetic data generation plays an ever-increasing role in NLP research, we expect this work to be a stepping stone in improving its utility. We conclude this paper with some recommendations on how to generate high(er)-fidelity synthetic data for specific tasks.

cs.CL↗

arXiv2023

Language Model Decoding as Likelihood-Utility Alignment

A critical component of a successful language generation pipeline is the decoding algorithm. However, the general principles that should guide the choice of a decoding algorithm remain unclear. Previous works only compare decoding algorithms in narrow scenarios, and their findings do not generalize across tasks. We argue that the misalignment between the model's likelihood and the task-specific notion of utility is the key factor to understanding the effectiveness of decoding algorithms. To structure the discussion, we introduce a taxonomy of misalignment mitigation strategies (MMSs), providing a unifying view of decoding as a tool for alignment. The MMS taxonomy groups decoding algorithms based on their implicit assumptions about likelihood--utility misalignment, yielding general statements about their applicability across tasks. Specifically, by analyzing the correlation between the likelihood and the utility of predictions across a diverse set of tasks, we provide empirical evidence supporting the proposed taxonomy and a set of principles to structure reasoning when choosing a decoding algorithm. Crucially, our analysis is the first to relate likelihood-based decoding algorithms with algorithms that rely on external information, such as value-guided methods and prompting, and covers the most diverse set of tasks to date. Code, data, and models are available at https://github.com/epfl-dlab/understanding-decoding.

cs.CL↗