Flow matching has shown strong promise as a continuous generative framework, but its standard formulation is not well-suited to reasoning problems. Conventional flow matching trains a model to transport noise directly toward a single final target. For tasks that require multi-step reasoning, this direct path is often too difficult: the final answer may be reachable only through a sequence of intermediate computations or transformations. Inspired by how autoregressive language models address this challenge through chain-of-thought supervision, we introduce ReasonFlow, a framework that brings the same scaffolding principle to flow matching. Instead of supervising the flow with only a final target, ReasonFlow constructs a sequence of CoT-induced anchor points derived from intermediate reasoning prefixes. During training, the clean target moves along a piecewise-linear path through these anchors before endpoint answer readout. This turns reasoning supervision from an output sequence into a trajectory-shaping signal for flow matching. Across arithmetic, chemistry, and logical-reasoning tasks, ReasonFlow substantially improves over no-CoT flow-matching baselines.

Chemistry literature synthesis often requires assembling specific findings scattered across many publications, yet existing literature-search systems primarily return ranked document lists. As a result, scientists and AI agents need to locate relevant information, trace each finding to its source, and assemble cross-paper answers manually. We present AskChem, a claim-centered infrastructure that changes the retrieval unit for cross-paper chemistry search from the paper to the source-linked claim. Available source text is converted into atomic, typed claims, each carrying a source DOI and a verbatim quote or an explicit evidence locator. Over this shared claim store, AskChem exposes complementary operational structures for search and synthesis: a corpus-induced faceted taxonomy stabilized for hierarchical retrieval and browsing, an automatically extracted evidence graph linking claims through typed relations, and an exploratory living taxonomy that situates indexed papers under scientific principles. At the frozen index snapshot, AskChem contains 2.4M claims from 147K papers, 30% processed from full text and 70% from abstracts only. The system provides a web interface, as well as REST, SDK, and MCP access for AI agents. On a 30-question AskChem-Bench pilot focused on citation traceability and paper relevance rather than factual-answer accuracy, grounding a GPT-5.5 reader in AskChem yields 100% resolvable DOIs, compared with 88.3% without retrieval, and the highest citation density among five tested systems.