Agentics 2.0: Logical Transduction Algebra for Agentic Data Workflows

Alfio Massimiliano Gliozzo; Junkyu Lee; Nahuel Defosse

cs.AIMar 4, 2026

Agentics 2.0: Logical Transduction Algebra for Agentic Data Workflows

Authors: Alfio Massimiliano Gliozzo, Junkyu Lee, Nahuel Defosse

Organizations: IBM

Abstract

Agentic AI is rapidly transitioning from research prototypes to enterprise deployments, where requirements extend to meet the software quality attributes of reliability, scalability, and observability beyond plausible text generation. We present Agentics 2.0, a lightweight, Python-native framework for building high-quality, structured, explainable, and type-safe agentic data workflows. At the core of Agentics 2.0, the logical transduction algebra formalizes a large language model inference call as a typed semantic transformation, which we call a transducible function that enforces schema validity and the locality of evidence. The transducible functions compose into larger programs via algebraically grounded operators and execute as stateless asynchronous calls in parallel in asynchronous Map-Reduce programs. The proposed framework provides semantic reliability through strong typing, semantic observability through evidence tracing between slots of the input and output types, and scalability through stateless parallel execution. We instantiate reusable design patterns and evaluate the programs in Agentics 2.0 on challenging benchmarks, including DiscoveryBench for data-driven discovery and Archer for NL-to-SQL semantic parsing, demonstrating state-of-the-art performance.

Figures & tables

Panel of Figure 1 — Figure 1: Overall aggregated scores. All figures show the aggregated scores over 10 problem domains, evaluating agentics-agg (ag-agg), agentics-both (ag-both), agentics-react (ag-react), and baseline-react (bl-react) using gemini-3-flash-previuew (gemini3flash) and gpt-4.1 (gpt4.1) models.

CSV File	Rows	Cols
archaeology/capital.csv	400	14
archaeology/time_series_data.csv	3600	37
archaeology/pollen_openness_score_…csv	4084	3
intro-path-plants/temporal_trends_contingency_table.csv	12	3
intro-path-plants/invasion_success_pathways.csv	81	19
intro-path-plants/invaded_niche_pathways.csv	632	25

Table 1: Size of CSV Files. The number of rows and columns of the CSV files in each dataset.

Panel of Figure 2 — Figure 2: Aggregated scores per dataset. All figures show the aggregated hypothesis matching scores per each dataset, and show the results from four algorithm configurations agentics-agg (ag-agg), agentics-both (ag-both), agentics-react (ag-react), and baseline-react (bl-react).

Panel of Figure 3 — Figure 3: Aggregated scores per question types. All figures show the aggregated scores for three question types: context questions, relationship questions, and variable questions.

Figure 4: Archer English Devset Execution Match. The figure shows the aggregated execution match score on the English devest, and shows the results from Agentics 2.0 implementation in blue bars and the Archer leaderboard results in orange bars.

Reasoning	Model	EX	F1	Count
A	Gemini-3-flash	0.875	0.875	24
A	GPT-o3	0.833	0.872	24
A C	Gemini-3-flash	0.393	0.474	28
A C	GPT-o3	0.607	0.674	28
A H	Gemini-3-flash	0.682	0.808	22
A H	GPT-o3	0.591	0.682	22

Table 2: Archer NL to SQL by Reasoning Type. The score shows the result from the reasoning-validation agent using two best-performing models, Gemini-3-flash and GPT-O3. A, C, and H refer to arithematic, commonsense, and hypothetical reasoning, respectively.

Agentic Workflows

Agentic workflows coordinate AI agents, tools, and humans across multi-step tasks, emphasizing reliable execution, efficiency, and oversight.

See all 218 papers

May 4, 2026

AAFLOW: Scalable Patterns for Agentic AI Workflows

Agentic workflows in large language model systems integrate retrieval, reasoning, and memory, but existing frameworks suffer from scalability and reproducibility limitations due to fragmented data orchestration, serialization overhead, and non-deterministic execution. Although these frameworks increase flexibility, they don't have a formal execution model that adheres to the principles of high-performance computing. We introduce AAFLOW, a unified distributed runtime that creates communication-efficient execution plans by modeling agentic workflows as an operator abstraction. Using Apache Arrow and Cylon, AAFLOW creates a zero-copy data plane that allows direct interoperability between preprocessing, embedding, and vector retrieval without the need for serialization overhead. To lower coordination costs, it uses resource-deterministic scheduling and asynchronous batching. While retaining comparable LLM generation throughput, experimental results demonstrate up to 4.64 times pipeline speedup and 2.8 times gains in embedding and upsert phases. Rather than LLM inference acceleration, these advantages result from enhanced data flow, batching, and communication efficiency.

Arup Kumar Sarker, Mills Staylor, Aymen Alsaadi +3

Department of Computer Science, Biocomplexity Institute and Initiative University of Virginia Charlottesville, VA, USA · Department of Computer Science, Rutgers University New Brunswick, NJ, USA · Biocomplexity Institute and Initiative University of Virginia Charlottesville, VA, USA +1

LLM Agent Orchestration Agentic Workflows

Ask PDF

Feb 18, 2026

TabAgent: Distilling LLM Agent Components into Cheap Classifiers

Agentic systems, AI architectures that autonomously execute multi-step workflows to achieve complex goals, are often built using repeated large language model (LLM) calls for closed-set decision tasks such as routing, shortlisting, gating, and verification. While convenient, this design makes deployments slow and expensive due to cumulative latency and token usage. We propose TabAgent, a framework for replacing generative decision components in closed-set selection tasks with a compact textual-tabular classifier trained on execution traces. TabAgent (i) extracts structured schema, state, and dependency features from trajectories (TabSchema), (ii) augments coverage with schema-aligned synthetic supervision (TabSynth), and (iii) scores candidates with a lightweight classifier (TabHead). On the long-horizon AppWorld benchmark, TabAgent maintains task-level success while eliminating shortlist-time LLM calls, reducing latency by approximately 95% and inference cost by 85-91%. Beyond tool shortlisting, TabAgent generalizes to other agentic decision heads, establishing a paradigm for learned discriminative replacements of generative bottlenecks in production agent architectures.

Ido Levy, Segev Shlomov, Avi Yaeli +5

IBM Research, Haifa, Israel

LLM Agent Workflow Optimization AI Agent Architecture LLM Decision-Making Classification

Ask PDF

May 5, 2026

Agentic-imodels: Evolving agentic interpretability tools via autoresearch

Agentic data science (ADS) systems are rapidly improving their capability to autonomously analyze, fit, and interpret data, potentially moving towards a future where agents conduct the vast majority of data-science work. However, current ADS systems use statistical tools designed to be interpretable by humans, rather than interpretable by agents. To address this, we introduce Agentic-imodels, an agentic autoresearch loop that evolves data-science tools designed to be interpretable by agents. Specifically, it develops a library of scikit-learn-compatible regressors for tabular data that are optimized for both predictive performance and a novel LLM-based interpretability metric. The metric measures a suite of LLM-graded tests that probe whether a fitted model's string representation is "simulatable" by an LLM, i.e. whether the LLM can answer questions about the model's behavior by reading its string output alone. We find that the evolved models jointly improve predictive performance and agent-facing interpretability, generalizing to new datasets and new interpretability tests. Furthermore, these evolved models improve downstream end-to-end ADS, increasing performance for Copilot CLI, Claude Code, and Codex on the BLADE benchmark by up to 73%

Chandan Singh, Yan Shuo Tan, Weijia Xu +4

Microsoft Research · National University of Singapore

Interpretable ML Automated Machine Learning Tabular Regression Data Analysis Agents

Ask PDF

Abstract

Agentic Workflows

Explore similar work

AAFLOW: Scalable Patterns for Agentic AI Workflows

TabAgent: Distilling LLM Agent Components into Cheap Classifiers

Agentic-imodels: Evolving agentic interpretability tools via autoresearch