Tag
This paper proposes a dependency-graph framework to formalize compositional reasoning in language models and evaluates the impact of reinforcement learning post-training, finding an asymmetry where composed-skill training transfers more readily to decomposed tasks than vice versa.
This paper proposes a dependency-aware trajectory refinement method for efficient multi-turn agent fine-tuning, improving accuracy and reducing inference costs.
This paper presents HSDN, a hierarchical semantic distillation network for compressing chain-of-thought reasoning in financial tasks, achieving 91.0% accuracy with 68.4% compression on the AFAC2025 benchmark.
aimake is an incremental build system for AI/ML pipelines that uses content-based fingerprints to rebuild only changed steps, optimizing expensive pipeline reruns.
This paper introduces dependency-guided rollback repair for memory-augmented agents, a method that builds a typed memory-to-action graph from runtime provenance to selectively undo faulty memory effects while preserving benign state, achieving strong recovery on benchmarks.
CodeFlow is a browser-based tool that visualizes code dependencies from a GitHub URL or local folder, providing interactive graphs, security scanning, pattern detection, and health scores without uploading code.
Frond is a frontend runtime that manages the dependency graph of React apps, handling service lifecycles and cleanup automatically.
TheoremGraph is a unified statement-level dependency graph that spans both informal mathematics (arXiv papers) and formal mathematics (Lean projects), using semantic embeddings to bridge the gap between them. The authors provide datasets, extractors, and APIs to support mathematical search and retrieval.
Archex is a new open-source code RAG tool that improves retrieval by combining hybrid search (BM25F + dense embeddings), cross-encoder reranking, and dependency-graph expansion, achieving much higher recall and token efficiency than pure embeddings-based approaches.
Graperoot is an MCP-native tool that builds a dependency graph of a codebase to avoid unnecessary file re-reading, saving users significant costs—over $150k collectively—and is free for any CLI or IDE supporting MCP.
Introduces ModSleuth, an agentic system that recursively reconstructs large-scale dependency graphs for LLM development by analyzing public artifacts, revealing multi-hop license obligations and documentation inconsistencies.
Repowise is an open-source tool that indexes codebases into five intelligence layers—dependency graph, git history, auto-generated docs, architectural decisions, and code health—and exposes them to AI coding agents via MCP tools for more accurate context and fewer tool calls.
A fully open-source codebase intelligence engine called SocratiCode helps AI navigate repositories using semantic search, dependency graphs, impact analysis, and shared indexes without vendor lock-in.
Recommend using both codegraph and graphify together with vibe coding. Codegraph analyzes dependency relationships between code (AST), graphify links concepts and code (AST + LLM semantics).
The author reflects on the challenge of untangling a complex dependency graph for their Intertwingler application server, aiming to achieve dense hypermedia to reduce the need for excessive reading and writing, while criticizing AI for generating too much content.
This paper introduces Grounded Continuation, a linear-time runtime verifier for LLM conversations that maintains an explicit dependency graph to detect whether a next utterance is supported by prior conversation, achieving accuracy gains over baselines on benchmarks including LongMemEval and LoCoMo.
MathAtlas is a large-scale benchmark for autoformalization of graduate-level mathematics, containing ~52k theorems and definitions extracted from 103 textbooks, with a mathematical dependency graph of ~178k relations. Experiments show state-of-the-art models achieve at most 9.8% correctness, highlighting the difficulty.