{"arxiv_id":"2608.18066","incoming":{"papers":1,"sentences":1,"groups":[],"unlabelled_sentences":1},"outgoing":{"papers":18,"sentences":27,"groups":[{"kind":"method_or_tool","label":"Uses the method or tool of","papers":1,"sentences":1,"quotes":[{"arxiv_id":"2509.26506","title":null,"url":"/paper/2509.26506","section":"Experiment Setting","sentence":"We use the baseline agent harness used in [ref] for WebArena and VisualWebArena, and the baseline agent harness used in [ref] for SCUBA.","intent_p":1.0,"contests_p":null}]},{"kind":"supports_this_paper","label":"Cites as support","papers":2,"sentences":3,"quotes":[{"arxiv_id":"2406.12045","title":"$\\tau$-bench: A Benchmark for Tool-Agent-User Interaction in Real-World Domains","url":"/paper/2406.12045","section":"Experiment Setting","sentence":"While this is permissible for early exploration in this field and for cost considerations, recent work points out that agent performance can vary across runs [ref] , and hence a comprehensive re-evaluation over multiple runs is needed.","intent_p":1.0,"contests_p":null},{"arxiv_id":"2602.16666","title":"Towards a Science of AI Agent Reliability","url":"/paper/2602.16666","section":"Experiment Setting","sentence":"While this is permissible for early exploration in this field and for cost considerations, recent work points out that agent performance can vary across runs [ref] , and hence a comprehensive re-evaluation over multiple runs is needed.","intent_p":1.0,"contests_p":null}]},{"kind":"background","label":"Cites as background","papers":13,"sentences":13,"quotes":[{"arxiv_id":"2302.04761","title":"Toolformer: Language Models Can Teach Themselves to Use Tools","url":"/paper/2302.04761","section":"Related Works","sentence":"The research community has developed diverse benchmarks for evaluating agent capabilities, in domains such as tool use [ref] , computer use [ref] , web browsing [ref] .","intent_p":0.97,"contests_p":null},{"arxiv_id":"cs/0309048","title":"Goedel Machines: Self-Referential Universal Problem Solvers Making Provably Optimal Self-Improvements","url":"/paper/cs/0309048","section":"Introduction","sentence":"The development of self-improving agents [ref] —systems designed to learn from past experiences and autonomously refine their performance over time—has emerged as a highly promising research direction.","intent_p":1.0,"contests_p":null},{"arxiv_id":"2310.08560","title":"MemGPT: Towards LLMs as Operating Systems","url":"/paper/2310.08560","section":"Related Works","sentence":"This work focuses on non-parametric, textual memory, which has gained significant traction recently [ref] .","intent_p":0.85,"contests_p":null},{"arxiv_id":"2401.01335","title":"Self-Play Fine-Tuning Converts Weak Language Models to Strong Language Models","url":"/paper/2401.01335","section":"Related Works","sentence":"More broadly, continual learning and self-improvement (or “self-play”) have long been studied prior to the inference-only paradigm, including in supervised fine-tuning [ref] and reinforcement learning settings [ref] .","intent_p":0.99,"contests_p":null},{"arxiv_id":"2510.04374","title":"GDPval: Evaluating AI Model Performance on Real-World Economically Valuable Tasks","url":"/paper/2510.04374","section":"Related Works","sentence":"As agent capabilities continue to grow, newer benchmarks target tasks with real-world economical values [ref] , including coding [ref] , customer service [ref] , customer relationship management [ref] , and enterprise operations [ref] .","intent_p":0.82,"contests_p":null}]}],"unlabelled_sentences":10},"claims":[],"thresholds":{"intent_p":0.8,"supports_p":0.95,"contests_p":0.7},"method":"Citing sentences come from each paper's LaTeX source; every \\cite marks the sentence and the works it cites. Pith's classifier typed each sentence. A label is shown at intent probability 0.8 or more (0.95 for support), a contest at 0.7 or more. Measured accuracy against two blind model families is at https://pith.science/accuracy. The quoted sentences are the citing authors' own.","held_up":null,"_links":{"html":"https://pith.science/paper/2608.18066"}}