Friday, July 31, 2026

1. Beyond Sentiment: Structured Information Extraction from Financial News
   T2arXiv AI/CL/LG（论文）

   Financial sentiment analysis has become standard for news-driven stock prediction, but it reduces rich news to a single polarity score. Researchers propose a structured information extraction framework using LLaMA-3.1-70B to extract six semantic dimensions from 41,618 news-stock pairs in the FNSPID dataset. Experiments show FinBERT sentiment features perform well under nonlinear models (F1=0.576) but weakly under linear models (F1=0.230). LLM-extracted structured features show a 53.5% systematic disagreement with sentiment, and combining both yields F1=0.600, significantly better than either alone. Ablation confirms non-sentiment dimensions independently add ΔF1=+0.019 beyond FinBERT. The results demonstrate sentiment-semantics decoupling is systematic and exploitable, opening new directions for multi-dimensional financial NLP.

—

1. AI Hedge Fund 'Situational Awareness' Liquidates Public Portfolio but Retains Anthropic Stake
   T2TechCrunch：AI（RSS）

   The AI hedge fund 'Situational Awareness', founded by a former OpenAI researcher, was forced to unwind its public equities after leveraged public bets plummeted. However, the fund still holds shares in AI startup Anthropic, retaining key assets and potential recovery opportunities. This move draws attention to risk management in AI-focused hedge funds.

2. Reddit Posts Solid Quarter, but AI Impact and Google Ties Stir Market Concerns
   T2TechCrunch：AI（RSS）

   Reddit's latest quarterly earnings show strong revenue and user growth, but the market is uncertain about its search traffic partnership with Google and the broader AI reshaping of the web. Investors worry AI-driven information aggregation may erode Reddit's community value, putting pressure on the stock.

3. Judge Rules Trump Admin Lacks Evidence to Label Anthropic a Supply Chain Risk
   T2TechCrunch：AI（RSS）

   A federal judge ruled that the Trump administration failed to provide sufficient evidence to label AI company Anthropic a supply chain risk, casting doubt on the government's ban on its AI technology. The ruling could reshape U.S. AI regulatory policy and spark debate on balancing national security with tech innovation.

4. Smallest.ai Raises $13M to Build Ultra-Fast Voice AI That Sounds Genuinely Human
   T2TechCrunch：AI（RSS）

   Smallest.ai has raised $13 million in funding to develop ultra-fast voice AI that sounds genuinely human, with the ambitious goal of enabling AI phone calls to pass the Turing test. The startup focuses on reducing latency and making conversations feel natural, which is critical for real-time telephony applications. With this capital, the company plans to expand its engineering team, refine its proprietary voice models, and accelerate go-to-market efforts across industries such as customer service, telemarketing, and virtual assistants. The technology could significantly improve user experience in automated voice interactions, making them indistinguishable from human conversations. This funding round reflects growing investor interest in voice AI as a key frontier in artificial intelligence, particularly in applications where realism and speed are paramount. As competition intensifies, Smallest.ai aims to differentiate itself by building models that not only respond quickly but also convey human nuances such as tone, emotion, and conversational rhythm.

—

—