Related papers: Active Evaluation of General Agents: Problem Defin…
Driving progress of AI models and agents requires comparing their performance on standardized benchmarks; for general agents, individual performances must be aggregated across a potentially wide variety of different tasks. In this paper, we…
Evaluating AI agents on comprehensive benchmarks is expensive because each evaluation requires interactive rollouts with tool use and multi-step reasoning. We study whether small task subsets can preserve agent rankings at substantially…
Unlike traditional automation tools or static LLM-based systems, agents combine decision-making and tool utilization to accomplish complex tasks, showing great potential in software engineering. However, existing studies largely focus on…
Load balancing across a networked environment is a monotonous job. Moreover, if the job to be distributed is a constraint satisfying one, the distribution of load demands core intelligence. This paper proposes parallel processing through…
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…
In this work, we explore the Large Language Model (LLM) agent reviewer dynamics in an Elo-ranked review system using real-world conference paper submissions. Multiple LLM agent reviewers with different personas are engage in multi round…
Evaluating performance across optimization algorithms on many problems presents a complex challenge due to the diversity of numerical scales involved. Traditional data processing methods, such as hypothesis testing and Bayesian inference,…
We present an alternative methodology for the analysis of algorithms, based on the concept of expected discounted reward. This methodology naturally handles algorithms that do not always terminate, so it can (theoretically) be used with…
As the field progresses toward Artificial General Intelligence (AGI), there is a pressing need for more comprehensive and insightful evaluation frameworks that go beyond aggregate performance metrics. This paper introduces a unified rating…
Background: The surge in single-cell omics data exposes limitations in traditional, manually defined analysis workflows. AI agents offer a paradigm shift, enabling adaptive planning, executable code generation, traceable decisions, and…
AI agents -- systems that combine foundation models with reasoning, planning, memory, and tool use -- are rapidly becoming a practical interface between natural-language intent and real-world computation. This survey synthesizes the…
The Elo rating system has been used world wide for individual sports and team sports, as exemplified by the European Go Federation (EGF), International Chess Federation (FIDE), International Federation of Association Football (FIFA), and…
We argue that many general evaluation problems can be viewed through the lens of voting theory. Each task is interpreted as a separate voter, which requires only ordinal rankings or pairwise comparisons of agents to produce an overall…
Rating strategies in a game is an important area of research in game theory and artificial intelligence, and can be applied to any real-world competitive or cooperative setting. Traditionally, only transitive dependencies between strategies…
The increasing adoption of foundation models as agents across diverse domains necessitates a robust evaluation framework. Current methods, such as LLM-as-a-Judge, focus only on final outputs, overlooking the step-by-step reasoning that…
Evaluating learned robot control policies to determine their physical task-level capabilities costs experimenter time and effort. The growing number of policies and tasks exacerbates this issue. It is impractical to test every policy on…
Agentic search -- the task of training agents that iteratively reason, issue queries, and synthesize retrieved information to answer complex questions -- has achieved remarkable progress through reinforcement learning (RL). However,…
Recent advances in agentic AI have shifted the focus from standalone Large Language Models (LLMs) to integrated systems that combine LLMs with tools, memory, and other agents to perform complex tasks. These multi-agent architectures enable…
Web-based 'deep research' agents aim to solve complex question - answering tasks through long-horizon interactions with online tools. These tasks remain challenging, as the underlying language models are often not optimized for long-horizon…
The task of building general agents that perform well over a wide range of tasks has been an important goal in reinforcement learning since its inception. The problem has been subject of research of a large body of work, with performance…