Daily Issue
Vol. I — No. 25
08 · 06
Monday, 8 June 2026
Generated 2026-06-08 13:53
google/gemini-2.5-flash-lite-preview-09-2025
陪着你做前端 — 萌芽 43 items · 4 sections
§ 0

The Morning

Local weather 1
This morning in
London
Overcast
Today's range
16.8°↓13.0°
currently 14.8°
Feels
13.4°
Rain
100%
Wind
9 km/h
Humid
70%
Rise
04:45
Set
21:15
§ I

US Stocks

Pre-market signal radar 12
US pre-market radar
premarket 2026-06-08
0 Bullish
0 Bearish
12 Neutral
Sector Tape
Hyperscale Cloud 4 names
42 Top: GOOGL · Neutral · RS -1.0% Bullish 0 / Bearish 0 / 5d -6.3%
Compute Mining 4 names
46 Top: IREN · Neutral · RS +0.6% Bullish 0 / Bearish 0 / 5d -8.9%
Energy Infrastructure 1 names
46 Top: VST · Neutral · RS -8.3% Bullish 0 / Bearish 0 / 5d -7.2%
Networking Equipment 4 names
46 Top: CIEN · Neutral · RS -3.9% Bullish 0 / Bearish 0 / 5d -9.5%
Manufacturing 4 names
52 Top: SANM · Neutral · RS -0.6% Bullish 0 / Bearish 0 / 5d -2.7%
Servers and Thermal Management 2 names
48 Top: DELL · Neutral · RS -0.3% Bullish 0 / Bearish 0 / 5d -5.5%
Battery and Energy Storage 3 names
49 Top: FLNC · Neutral · RS +2.9% Bullish 0 / Bearish 0 / 5d -1.9%
Foundry 2 names
51 Top: INTC · Neutral · RS -2.1% Bullish 0 / Bearish 0 / 5d -7.2%
Ticker Setup Move Score Evidence Quality
FLNC Fluence Energy Battery and Energy Storage
Neutral Gap up + news Low confidence
+7.0% $24.52 5d +21.4%
65 sector negative RS +26.2%

Watchlist item from +7.0% vs previous close, negative sector tape, 3 recent headline(s).

Fluence Energy (FLNC) stock trades up, here is why - MSN Needs fresh price/news confirmation before becoming an actionable setup.
quote: delayed fallback news: fresh financials: fresh news: 3
SLDP Solid Power Battery and Energy Storage
Neutral Baseline watch Low confidence
+1.3% $2.98 5d -11.2%
40 sector negative RS -6.3%

Watchlist item from +1.3% vs previous close, negative sector tape.

Needs fresh price/news confirmation before becoming an actionable setup.
quote: delayed fallback news: sparse financials: fresh news: 0
AMZN Amazon Hyperscale Cloud
Neutral News watch Low confidence
+0.6% $247.45 5d -9.1%
42 sector negative RS -3.8%

Watchlist item from negative sector tape, 3 recent headline(s).

AMZN Stock Quote Price and Forecast - CNN Needs fresh price/news confirmation before becoming an actionable setup.
quote: delayed fallback news: fresh financials: fresh news: 3
SANM Sanmina Manufacturing
Neutral Gap up + news Low confidence
+1.9% $256.96 5d -3.0%
56 sector negative RS -0.8%

Watchlist item from +1.9% vs previous close, negative sector tape, 3 recent headline(s).

Sanmina Corporation $SANM Holdings Increased by Vestcor Inc - MarketBeat Needs fresh price/news confirmation before becoming an actionable setup.
quote: delayed fallback news: fresh financials: fresh news: 3
HUT Hut 8 Compute Mining
Neutral Gap up + news Low confidence
+2.3% $114.80 5d -10.1%
45 sector negative RS -0.5%

Watchlist item from +2.3% vs previous close, negative sector tape, 3 recent headline(s).

Why is HUT stock sliding over 2% premarket today? - MSN Needs fresh price/news confirmation before becoming an actionable setup.
quote: delayed fallback news: fresh financials: fresh news: 3
APH Amphenol Networking Equipment
Neutral Gap up + news Low confidence
+3.5% $143.61 5d -6.7%
46 sector negative RS -1.1%

Watchlist item from +3.5% vs previous close, negative sector tape, 3 recent headline(s).

Is Amphenol stock outperforming the Nasdaq? - MSN Needs fresh price/news confirmation before becoming an actionable setup.
quote: delayed fallback news: fresh financials: fresh news: 3
quotes: nasdaq 24 24/24news: google_news_rss 22 22/24filings: sec 24 24/24, fallback 24

Generated from public market data and news for research and education. Not financial advice; data may be delayed, incomplete, or wrong.

§ II

From the arXiv

arXiv preprints 10 of 20
cs.LGarxiv:2606.07367v1Lead article

Self-evolving LLM agents with in-distribution Optimization

Yudi Zhang, Meng Fang, Zhenfang Chen, Mykola Pechenizkiy

he paper introduces **Q-Evolve**, a self-evolving framework for LLM agents designed to overcome sparse reward challenges in long-horizon decision-making. It unifies automatic process-reward labeling and policy learning using an in-distribution reinforcement learning approach. The core method learns a stable critic from a hybrid dataset using a weighted Implicit Q-Learning objective, which then generates dense, step-wise process rewards via advantage estimation for improved supervision.

Read abstract →Full PDFLINNET · SCORE10RELEVANCE
Comparison of existing methods. Left: Existing PRM methods rely on costly manual labels or search-based rollouts requiring discrete states, often failing due to distribution shifts between PRM training and policy improvement. Upper Mid: Most online RL does not address episodic sparse rewards. Bottom Mid: Our framework utilizes a hybrid off-policy dataset (expert + agents’ interaction data) to derive rewards via Bellman backups. By co-evolving process reward supervision and policy improvement within a shared in-distribution loop, the agent achieves stable self-evolution. Right: A visualization of performance vs environment steps required for collecting data.
Comparison of existing methods. Left: Existing PRM methods rely on costly manual labels or search-based rollouts requiring discrete states, often failing due to distribution shifts between PRM training and policy improvement. Upper Mid: Most online RL does not address episodic sp…
State CoT: A transition diagram illustrating the discrete reasoning states and meta-cognitive actions within a trajectory.
State CoT: A transition diagram illustrating the discrete reasoning states and meta-cognitive actions within a trajectory.
cs.AIarxiv:2606.07410v1

A Comprehensive Anatomy of Human and DeepSeek-R1 LLM Mathematical Reasoning

Yuxiang Chen, Jun Wang

This paper comprehensively compares the mathematical reasoning steps of the DeepSeek-R1 LLM and humans on AIME 2025 problems, categorizing 10,247 steps. The core finding is a structural difference: human reasoning is compact, while the LLM exhibits "topologica…

abstractpdfLINNET · SCORE9RELEVANCE
cs.AIarxiv:2606.07462v1

Act As a Real Researcher: A Suite of Benchmarks Evaluating Frontier LLMs and Agentic Harnesses in Research Lifecycle

Jiayu Wang, Weijiang Lv et al.

This paper introduces the **AARR (Act As a Real Researcher) benchmark series** to evaluate frontier LLMs and agents on the nuanced professionalism and thoroughness required in real research, moving beyond simple macro-level execution. The first installment, **…

abstractpdfLINNET · SCORE9RELEVANCE
Overview of the AARRI-Bench Pipeline. The benchmark is constructed through a three-stage human-in-the-loop workflow with two-dimensional task categorization across task scenarios and agent scope levels. Tasks are evaluated under the Harbor framework with standardized environments, multiple agent harnesses and models, and both coarse-grained and fine-grained metrics.
Overview of the AARRI-Bench Pipeline. The benchmark is constructed through a three-stage human-in-the-loop workflow with two-dimensional task categorization across task scenarios and agent scope level…
The illustration for the Qianfan Agent Foundry.
The illustration for the Qianfan Agent Foundry.
cs.AIarxiv:2606.07299v1

DuMate-DeepResearch: An Auditable Multi-Agent System with Recursive Search and Rubric-Grounded Reasoning

Lingyong Yan, Can Xu et al.

DuMate-DeepResearch is a multi-agent framework designed to overcome limitations in current Deep Research (DR) systems, specifically concerning long-horizon planning, task decomposition, and auditability. It achieves this by decoupling the Agent Core (handling …

abstractpdfLINNET · SCORE9RELEVANCE
cs.AIarxiv:2606.07489v1

How AI Agents Reshape Knowledge Work: Autonomy, Efficiency, and Scope

Jeremy Yang, Kate Zyskowski et al.

This paper investigates how autonomous AI agents transform knowledge work by analyzing production data comparing Perplexity's Search and Computer products. The core finding is that the autonomous Computer product significantly accelerates task completion (26 m…

abstractpdfLINNET · SCORE9RELEVANCE
AI product progression by autonomy and workflow-context integration. Perplexity’s Search represents the baseline for information retrieval and synthesis; Comet Assistant introduces deeper context integration and execution on top of an interactive browser interface; Computer combines long-horizon asynchronous execution with even deeper and broader context integration as an agent orchestrator.
AI product progression by autonomy and workflow-context integration. Perplexity’s Search represents the baseline for information retrieval and synthesis; Comet Assistant introduces deeper context inte…
№06
cs.AI
9

Online Pandora's Box for Contextual LLM Cascading

Alexandre Belloni, Yan Chen et al.

This paper introduces the **Online Pandora's Box for Contextual LLM Cascading**, an adaptive framework for sequentially querying and selecting among LLM APIs based on request conte…

abstractpdf
№07
cs.AI
9

Socratic-SWE: Self-Evolving Coding Agents via Trace-Derived Agent Skills

Chuan Xiao, Zhengbo Jiao et al.

Socratic-SWE is a closed-loop framework that enables self-evolving software engineering agents by leveraging their own historical solving traces. It distills these traces into stru…

abstractpdf
№08
cs.AI
9

SV-Detect: AI-generated Text Detection with Steering Vectors

Mikhail Vishnyakov, Tatiana Gaintseva

SV-Detect detects AI-generated text by extracting "steering vectors" from a frozen language model's hidden layers, which define directions separating human and machine text. The me…

abstractpdf
№09
cs.AI
9

When Large Language Models Fail in Healthcare: Evaluating Sensitivity to Prompt Variations

Mahdi Alkaeed

This paper systematically evaluates the sensitivity of general and medical Large Language Models (LLMs) to prompt variations (natural and adversarial) using the MedMCQA benchmark. …

abstractpdf
№10
cs.CL
9

Agentopia: Long-Term Life Simulation and Learning in Agent Societies

Xintao Wang, Sirui Zheng et al.

Agentopia is a comprehensive framework designed for long-term life simulation of multi-agent societies, extending simulations from days to years. The core method involves simulatin…

abstractpdf
See all 20 papers →
§ III

The Town Square

Hacker News 2
143
pts
Top story
The OnlyFans Economy of American AI

The story likely explores the economic impact and dynamics of American AI, possibly drawing a provocative comparison to the content-driven, subscription-based "OnlyFans economy."

leoveanu.com7 Jun discuss on HN →
112
If LLMs Have Human-Like Attributes, Then So Does Age of Empires II
arxiv.org7 Jun
compiled overnight by google/gemini-2.5-flash-lite-preview-09-2025 · end of issue no. 25 · thank you for reading