Existing textual-gradient prompt optimisers rely on a single Large Language Model (LLM) to diagnose failures, propose edits, and judge the results, conflating critique and refinement into one model’s judgement. We introduce DialPO (Dialectical Prompt Optimisation), a multi-agent framework that replaces the single- model feedback loop with a structured de- bate: an Advocate defends what the prompt does well, a Critic attacks failures, cross- examination filters their arguments adversar- ially, a data-isolated Synthesis model drafts the revised prompt from surviving arguments alone, and a regression gate rejects edits that degrade batch accuracy. We provide a theo- retical framework identifying three precondi- tions under which multi-agent debate improves the feedback-reasoning bottleneck of textual- gradient methods, and show that DialPO in- stantiates them by construction. Across 12 benchmarks spanning classification, reasoning, summarisation, and question answering, Di- alPO achieves the highest scores across various Target models, winning or tying a majority of datasets in head-to-head comparisons against each baseline while using 8–18× fewer LLM calls than beam-search optimisers. Compo- nent ablations across five Target models reveal that DialPO’s advantage scales inversely with Target-model strength, clarifying when dialec- tical debate helps prompt optimisation.

Can an LLM read a 30+ page Standard Operating Procedure (SOP) and produce a team of agents that reliably execute it? Industrial SOPs bury workflow logic in prose, leave inter-step dependencies implicit, and scatter referenced entities across sections, forcing planners to parse, sequence, and tool-map within one context window. We argue that the missing primitive is procedural chunking: decomposing a document into self-contained executable units. Unlike RAG-style chunking, each chunk must be independently executable and chunks must jointly form a valid DAG encoding data and process dependencies, turning a prose SOP into a compact orchestrable workflow graph. We present ProChunker, a three-phase self-optimizing LLM pipeline (Boundary Detection, Metadata Extraction, Dependency Construction) with Stochastic Best-of-N prompt exploration guided by five custom metrics (Boundary Clarity, Entity Reference Coverage, Agent Self-Containedness Agreement, Step Coverage, Graph Validity); we then fine-tune Qwen3-1.7B on the pipeline’s own phase outputs, cutting latency by ∼27 min. Across three corpora (500 BioProBench protocols, six Industrial SOPs, 14 SOP-Bench documents), ProChunker improves the five-metric composite over the strongest baseline by +0.036 to +0.111 across corpora. Our study shows ProChunker increases agent construction success rate by +17% and resolves >90% of procedural ambiguity.

Conversational Agents have become ubiquitous across application domains, such as, shopping assistants, medical diagnosis, autonomous task planning etc. Users interacting with these agents often fail to understand how to start a conversation or what to ask next to obtain the desired information. To enable seamless and hassle-free user-agent interactions, we introduce Next Question Suggestions (NQS), which are essentially highly relevant follow-up question recommendations that act as conversation starters or discover-ability tools to capture non-trivial user intents, leading to more engaging conversations. Relying on LLMs for both response as well as NQS generation is a costly ask in latency-constrained commercial settings, with an added risk of handling potentially unsafe or unanswerable generated queries. A key component of building an efficient low-latency NQS experience is, therefore, retrieval (or embedding) models that fetch the most-relevant candidate questions from an offline pre-curated Question Bank (QB). Off-the-shelf embedding models cannot capture domain-specific nuances and more importantly the directionality inherent in follow-up question recommendations. In this work, we propose an end-to-end retrieval system, DIRECT that is optimized to model directional relevance. Given a user query, it produces a ranked list of highly relevant follow-up question recommendations within 1 sec. Our system also contains an LLM-as-a-judge component, tuned on proprietary user-agent interaction logs, to evaluate the end-to-end performance in terms of CTR.

While Large Language Models excel at reasoning and language understanding, they struggle with multi-step operational workflows requiring precise procedural adherence, which is fundamental for industrial automation. Existing SOP-guided agents assume well-defined procedures and structured APIs, failing to address enterprise realities like incomplete SOPs, dynamic web interfaces, and unpredictable document formats. We present Agent-Ops, an end-to-end multi-agent framework automating Standard Operating Procedures in e-commerce. Agent-Ops contributes: (1) SOP Groomer, a human-AI framework transforming ambiguous documentation into automation-ready specifications, improving accuracy by 13.2%, (2) WebAgent, achieving 91.3% task completion and 86.5% execution consistency through demonstration-based learning, and (3) a Document Verification Agent performing multi-lingual validation across tax invoices, certificates, and supply chain documents with 94.2% accuracy. Deployed across seven SOP categories in three geographic regions, Agent-Ops achieves 85-97% end-to-end accuracy while reducing case resolution from 30 to 5 minutes (83% reduction). Production deployment with over 1000 Account Managers validates that LLM-based agents achieve enterprise-grade reliability when augmented with robust web automation, comprehensive document understanding, and systematic SOP refinement.

Identifying preferences of customers in their shopping journey is a pivotal aspect in providing product recommendations. The task becomes increasingly challenging when there is a multi-turn conversation between the user and a shopping assistant chatbot. In this paper, we tackle a novel and complex problem of identifying customer preferences in the form of key-value filters on an e-commerce website in a multi-turn conversational setting. Existing systems specialize in extracting customer preferences from standalone customer queries which makes them unsuitable to multi-turn setup. We propose PEARL (Preference Extraction with ICL Augmentation and Retrieval with LLM Agents) that leverages collaborative LLM agents, generates in-context learning exemplars and dynamically retrieves relevant exemplars during inference time to extract customer preferences as a combination of key-value filters. Our experiments on proprietary and public datasets show that PEARL not only improves performance on exact match by ~10% compared to competitive LLM-based baselines but additionally improves inference latency by ~110%.