Industrial asset operations workflows are latency-sensitive because a single user<br>query may require coordination over sensor data, work orders, failure modes, fore-<br>casting tools, and domain-specific agents. We evaluate this problem on Asse-<br>tOpsBench (AOB), an industrial agent benchmark whose plan-execute pipeline<br>exposes repeated overhead from tool discovery, LLM planning, MCP tool exe-<br>cution, and final summarization. Existing LLM caching techniques such as KV-<br>cache reuse and embedding-based semantic caching were designed for chatbot<br>serving and break down when output validity depends on time, asset, or sen-<br>sor parameters. We propose two complementary optimization layers for AOB<br>plan-execute pipelines: a temporal semantic cache and a set of MCP workflow<br>optimizations combining disk-backed tool-discovery caching and dependency-<br>aware parallel step execution. MCP workflow optimizations corresponded to a<br>1.67× speedup and reduced median end-to-end latency by about 40.0% while the<br>temporal-cache benchmark achieved a median of 30.6× speedup on cache hits.<br>Beyond the speedup, our results expose a concrete failure mode of pure seman-<br>tic caching for parameter-rich industrial queries, providing a critical analysis of<br>how caching choices interact with evaluation correctness in MCP-backed agent<br>benchmarks.</p>\n","updatedAt":"2026-05-21T02:11:03.702Z","author":{"_id":"64c47f731d44fc06afc80953","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/UT2mHX2WuCm5Ws4rGKyCB.png","fullname":"Dhaval Patel","name":"DhavalPatel","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":9,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.8504114747047424},"editors":["DhavalPatel"],"editorAvatarUrls":["https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/UT2mHX2WuCm5Ws4rGKyCB.png"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2605.20630","authors":[{"_id":"6a0e6996164dbbc68a26c46a","name":"Alimurtaza Mustafa Merchant","hidden":false},{"_id":"6a0e6996164dbbc68a26c46b","name":"Krish Veera","hidden":false},{"_id":"6a0e6996164dbbc68a26c46c","name":"Sajal Kumar Goyla","hidden":false},{"_id":"6a0e6996164dbbc68a26c46d","name":"Shambhawi Bhure","hidden":false},{"_id":"6a0e6996164dbbc68a26c46e","name":"Dhaval Patel","hidden":false},{"_id":"6a0e6996164dbbc68a26c46f","name":"Kaoutar El Maghraoui","hidden":false}],"publishedAt":"2026-05-20T00:00:00.000Z","submittedOnDailyAt":"2026-05-21T00:00:00.000Z","title":"Evaluating Temporal Semantic Caching and Workflow Optimization in Agentic Plan-Execute Pipelines","submittedOnDailyBy":{"_id":"64c47f731d44fc06afc80953","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/UT2mHX2WuCm5Ws4rGKyCB.png","isPro":false,"fullname":"Dhaval Patel","user":"DhavalPatel","type":"user","name":"DhavalPatel"},"summary":"Industrial asset operations workflows are latency-sensitive because a single user query may require coordination over sensor data, work orders, failure modes, forecasting tools, and domain-specific agents. We evaluate this problem on AssetOpsBench (AOB), an industrial agent benchmark whose plan-execute pipeline exposes repeated overhead from tool discovery, LLM planning, MCP tool execution, and final summarization. Existing LLM caching techniques such as KV-cache reuse and embedding-based semantic caching were designed for chatbot serving and break down when output validity depends on time, asset, or sensor parameters. We propose two complementary optimization layers for AOB plan-execute pipelines: a temporal semantic cache and a set of MCP workflow optimizations combining disk-backed tool-discovery caching and dependency-aware parallel step execution. MCP workflow optimizations corresponded to a 1.67x speedup and reduced median end-to-end latency by about 40.0% while the temporal-cache benchmark achieved a median of 30.6x speedup on cache hits. Beyond the speedup, our results expose a concrete failure mode of pure semantic caching for parameter-rich industrial queries, providing a critical analysis of how caching choices interact with evaluation correctness in MCP-backed agent benchmarks.","upvotes":9,"discussionId":"6a0e6996164dbbc68a26c470","ai_summary":"Industrial asset operations workflows face latency challenges due to complex coordination needs, addressed through novel caching and workflow optimization techniques that improve execution speed while maintaining correctness in parameter-rich environments.","ai_keywords":["temporal semantic cache","MCP workflow optimizations","disk-backed tool-discovery caching","dependency-aware parallel step execution","plan-execute pipeline","AssetOpsBench","LLM caching","KV-cache reuse","embedding-based semantic caching"],"organization":{"_id":"616e7b1d75754a5d5fa455cf","name":"ibm","fullname":"IBM","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/637bfdf60dc13843b468ac20/9228luWRoGbZwKGxkOOsj.png"}},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"64c47f731d44fc06afc80953","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/UT2mHX2WuCm5Ws4rGKyCB.png","isPro":false,"fullname":"Dhaval Patel","user":"DhavalPatel","type":"user"},{"_id":"6915ff8130500b43788ae3ac","avatarUrl":"/avatars/4b94cbfb9785a5cc4f162d7af928e4c9.svg","isPro":false,"fullname":"Krish Veera","user":"krishveera14","type":"user"},{"_id":"68bdcec783b40d71e14a07df","avatarUrl":"/avatars/2d57dfd49e304646d80b81a915afb00d.svg","isPro":false,"fullname":"Alimurtaza Merchant","user":"alimurtaza0411","type":"user"},{"_id":"69b065faba86efefd91a7ce1","avatarUrl":"/avatars/e2e4eb75eb09956250f3b2a8fc07392d.svg","isPro":false,"fullname":"Sajal Kumar Goyla","user":"SajalGoyla","type":"user"},{"_id":"6917be5aab32be016a17c811","avatarUrl":"/avatars/4fe93710947a463ec77002869aa82ed1.svg","isPro":false,"fullname":"Madhav Tibrewal","user":"madhavtibrewal92","type":"user"},{"_id":"69b06604531cfd1c221ad8db","avatarUrl":"/avatars/cd34a3567ab20157e32682978cf59dc9.svg","isPro":false,"fullname":"Shambhawi Bhure","user":"shambhawibhure","type":"user"},{"_id":"662e745901e4fa6f0104a964","avatarUrl":"/avatars/c2f8f1d51e4a8ef2c5bd6c9d97b33cfc.svg","isPro":false,"fullname":"Deipey Paanchal","user":"deipeypaanchal","type":"user"},{"_id":"66bc79f085f17cea2c317811","avatarUrl":"/avatars/695b12231f8460c8f7b14382c9e3a995.svg","isPro":false,"fullname":"FayedHakim","user":"LLMProj","type":"user"},{"_id":"69a3f35b34ec7f83af7d67ad","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/noauth/EdXIApkxUX9flqqrBQ3_d.png","isPro":false,"fullname":"郭思宇","user":"songwe1xj","type":"user"}],"acceptLanguages":["en"],"dailyPaperRank":0,"organization":{"_id":"616e7b1d75754a5d5fa455cf","name":"ibm","fullname":"IBM","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/637bfdf60dc13843b468ac20/9228luWRoGbZwKGxkOOsj.png"},"markdownContentUrl":"https://huggingface.co/buckets/huggingchat/papers-content/resolve/2605/2605.20630.md"}">
Evaluating Temporal Semantic Caching and Workflow Optimization in Agentic Plan-Execute Pipelines
Abstract
Industrial asset operations workflows face latency challenges due to complex coordination needs, addressed through novel caching and workflow optimization techniques that improve execution speed while maintaining correctness in parameter-rich environments.
AI-generated summary
Industrial asset operations workflows are latency-sensitive because a single user query may require coordination over sensor data, work orders, failure modes, forecasting tools, and domain-specific agents. We evaluate this problem on AssetOpsBench (AOB), an industrial agent benchmark whose plan-execute pipeline exposes repeated overhead from tool discovery, LLM planning, MCP tool execution, and final summarization. Existing LLM caching techniques such as KV-cache reuse and embedding-based semantic caching were designed for chatbot serving and break down when output validity depends on time, asset, or sensor parameters. We propose two complementary optimization layers for AOB plan-execute pipelines: a temporal semantic cache and a set of MCP workflow optimizations combining disk-backed tool-discovery caching and dependency-aware parallel step execution. MCP workflow optimizations corresponded to a 1.67x speedup and reduced median end-to-end latency by about 40.0% while the temporal-cache benchmark achieved a median of 30.6x speedup on cache hits. Beyond the speedup, our results expose a concrete failure mode of pure semantic caching for parameter-rich industrial queries, providing a critical analysis of how caching choices interact with evaluation correctness in MCP-backed agent benchmarks.
Community
Industrial asset operations workflows are latency-sensitive because a single user
query may require coordination over sensor data, work orders, failure modes, fore-
casting tools, and domain-specific agents. We evaluate this problem on Asse-
tOpsBench (AOB), an industrial agent benchmark whose plan-execute pipeline
exposes repeated overhead from tool discovery, LLM planning, MCP tool exe-
cution, and final summarization. Existing LLM caching techniques such as KV-
cache reuse and embedding-based semantic caching were designed for chatbot
serving and break down when output validity depends on time, asset, or sen-
sor parameters. We propose two complementary optimization layers for AOB
plan-execute pipelines: a temporal semantic cache and a set of MCP workflow
optimizations combining disk-backed tool-discovery caching and dependency-
aware parallel step execution. MCP workflow optimizations corresponded to a
1.67× speedup and reduced median end-to-end latency by about 40.0% while the
temporal-cache benchmark achieved a median of 30.6× speedup on cache hits.
Beyond the speedup, our results expose a concrete failure mode of pure seman-
tic caching for parameter-rich industrial queries, providing a critical analysis of
how caching choices interact with evaluation correctness in MCP-backed agent
benchmarks.
Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.
Tap or paste here to upload images
Cite arxiv.org/abs/2605.20630 in a model README.md to link it from this page.
Cite arxiv.org/abs/2605.20630 in a dataset README.md to link it from this page.
Cite arxiv.org/abs/2605.20630 in a Space README.md to link it from this page.
Discussion (0)
Sign in to join the discussion. Free account, 30 seconds — email code or GitHub.
Sign in →No comments yet. Sign in and be the first to say something.