# Вземане на решения

Вземане на решения, основано на данни, с финансови прозрения

4 публикации [Вижте всички етикети](https://beancount.io/bg/bean-labs/research-logs/tags) [Обратно към всички публикации](https://beancount.io/bg/bean-labs/research-logs)

1 юли 2026 г. 6 минути четене

## [Uncertainty-Aware Deferral for LLM Agents: When to Escalate from Small to Large Models](https://beancount.io/bg/bean-labs/research-logs/2026/07/01/redact-uncertainty-aware-deferral-llm-agents)

ReDAct defers from a small model to a large one only when perplexity signals uncertainty, cutting cost 64% at matching accuracy for agent workflows.

Mike Thrift

ai llm automation

2 юни 2026 г. 5 минути четене

## [InvestorBench: Qwen2.5-72B tops stocks at 46.15% CR](https://beancount.io/bg/bean-labs/research-logs/2026/06/02/investorbench-llm-agent-financial-decision-making)

InvestorBench: Qwen2.5-72B leads stock trading at 46.15% CR; finance-tuned Palmyra-Fin backfires on equities—size beats domain fine-tuning.

Mike Thrift

llm ai finance

10 май 2026 г. 6 минути четене

## [LATS: Езиков агент с дървовидно търсене — разсъждение, действие и планиране в една рамка](https://beancount.io/bg/bean-labs/research-logs/2026/05/10/lats-language-agent-tree-search-reasoning-acting-planning)

LATS обединява ReAct, Tree of Thoughts и Reflexion в една MCTS рамка, постигайки 92.7% pass\@1 на HumanEval с GPT-4.

Mike Thrift

ai llm machine-learning

27 април 2026 г. 7 минути четене

## [Tree of Thoughts: Deliberate Problem Solving with LLM Search](https://beancount.io/bg/bean-labs/research-logs/2026/04/27/tree-of-thoughts-deliberate-problem-solving)

Tree of Thoughts hits 74% on Game of 24 where GPT-4 chain-of-thought gets 4%, by searching and backtracking over steps, a pattern finance agents can borrow.

Mike Thrift

ai llm machine-learning

[Започни безплатно→](https://beancount.io/sign-up/?next=%2Fledger%2F) [Документация](https://beancount.io/bg/docs)

## Where to look next

- [Sitemap](https://beancount.io/sitemap.xml) - all pages with hreflang alternates
- [LLMs index](https://beancount.io/llms.txt) - curated links for AI agents (also /llms-full.txt for full docs)
- [Documentation](https://beancount.io/docs) - every doc is also available as text/markdown via .md suffix or Accept negotiation
- [REST API reference](https://beancount.io/docs/api-reference) - interactive endpoints, schemas, authentication, and code samples
- [OpenAPI spec](https://beancount.io/openapi.json) - OpenAPI 3.1 spec for this website's public content endpoints; the product contract for ledger operations is at https://beancount.io/api-gateway/v1/openapi.json
- [GraphQL](https://beancount.io/graphql) - public, read-only content API over the blog and research logs (no credentials, no ledger data): 12 query fields over 23 types
- [MCP server](https://beancount.io/mcp) - Model Context Protocol at https://beancount.io/api-gateway/mcp
