Are Online Skill and Memory Modules Always Worth Their Tokens? A Budget-Constrained Study of Web Agents
This study re-evaluates online augmentation modules for web agents under a fixed inference budget, finding that AWM, ASI, and ReasoningBank do not provide significant advantages over a token-matched vanilla baseline that uses the same budget for additional actor steps. The baseline matches or surpasses all augmentation methods in success rate while often using fewer total tokens. The effect extends to enterprise knowledge-work tasks, and run-to-run variance is highlighted as a crucial evaluation metric.
[2606.15017] Are Online Skill and Memory Modules Always Worth Their Tokens? A Budget-Constrained Study of Web Agents
[Submitted on 12 Jun 2026]
Title:Are Online Skill and Memory Modules Always Worth Their Tokens? A Budget-Constrained Study of Web Agents
View a PDF of the paper titled Are Online Skill and Memory Modules Always Worth Their Tokens? A Budget-Constrained Study of Web Agents, by Sina Hajimiri and 6 other authors
View PDF HTML (experimental)
Abstract:Online web agents often augment a base actor with memory, workflow, or skill modules. These modules can improve performance, but they also consume test-time tokens, a cost rarely reported alongside the actor's inference cost. We study online augmentation, where this overhead is paid on every task, and re-evaluate its benefits under a fixed total inference budget. We compare AWM, ASI, and ReasoningBank with a token-matched vanilla baseline that uses the same budget for additional actor steps. Across three WebArena domains and three models, Gemini 3 Flash, GPT-5.4-mini, and Qwen 3.6-27B, the vanilla baseline matches or surpasses all three augmentation methods in aggregate success rate while often using fewer total tokens. We observe a similar trend on WorkArena-L1 with Qwen 3.6-27B, indicating that the effect extends to enterprise knowledge-work tasks. Our results suggest that skills and workflow memory can be useful in specific domains, but their apparent gains often vanish against a budget-matched actor. We further show that run-to-run variance materially affects outcomes and should be reported as a core evaluation criterion for online web agents.
Subjects:
Computation and Language (cs.CL)
Cite as: arXiv:2606.15017 [cs.CL]
(or arXiv:2606.15017v1 [cs.CL] for this version)
https://doi.org/10.48550/arXiv.2606.15017
arXiv-issued DOI via DataCite (pending registration)
Submission history
From: Sina Hajimiri [view email] [v1] Fri, 12 Jun 2026 23:30:14 UTC (102 KB)
Full-text links:
Access Paper:
View a PDF of the paper titled Are Online Skill and Memory Modules Always Worth Their Tokens? A Budget-Constrained Study of Web Agents, by Sina Hajimiri and 6 other authors
View PDF
HTML (experimental)
TeX Source
view license
Current browse context:
cs.CL
new | recent | 2026-06
Change to browse by:
cs
References & Citations
NASA ADS
Google Scholar
Semantic Scholar
Loading...
Data provided by:
Bibliographic Tools
Bibliographic and Citation Tools
Bibliographic Explorer Toggle
Bibliographic Explorer (What is the Explorer?)
Connected Papers Toggle
Connected Papers (What is Connected Papers?)
Litmaps Toggle
Litmaps (What is Litmaps?)
scite.ai Toggle
scite Smart Citations (What are Smart Citations?)
Code, Data, Media
Code, Data and Media Associated with this Article
alphaXiv Toggle
alphaXiv (What is alphaXiv?)
Links to Code Toggle
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub Toggle
DagsHub (What is DagsHub?)
GotitPub Toggle
Gotit.pub (What is GotitPub?)
Huggingface Toggle
Hugging Face (What is Huggingface?)
ScienceCast Toggle
ScienceCast (What is ScienceCast?)
Demos
Demos
Replicate Toggle
Replicate (What is Replicate?)
Spaces Toggle
Hugging Face Spaces (What is Spaces?)
Spaces Toggle
TXYZ.AI (What is TXYZ.AI?)
Related Papers
Recommenders and Search Tools
Link to Influence Flower
Influence Flower (What are Influence Flowers?)
Core recommender toggle
CORE Recommender (What is CORE?)
Author
Venue
Institution
Topic
About arXivLabs
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)