.@POTUS & @VP Harris believe we need to build, build, build to lower housing costs. Today, the Administration is taking new actions to reduce barriers to build, making it easier & faster to use federal financing to build or rehab affordable housing. 🧵:
https://t.co/R7MI2C0AmA
With almost half of S&P500 companies reporting so far, the 9.8% earnings growth y/y this quarter (up 0.9% vs. initial expectations) looks pretty good.
That's until it gets compared to the 17% expected y/y earnings growth expected by 4Q24 and the 15% growth for '25 too.
Thread.
@BobEUnlimited Do you agree with @SteveMiran and @Nouriel's argument that the Treasury's adjusting of the maturity profile of its debt issuance through "Activist Treasury Issuance" (ATI) has dampened the restrictiveness of the FFR (by ~100bp)?
Current monetary policy has been in place for nearly 2 years, and yet the US economy as a whole hasn't slowed much.
The argument that policy is far too restrictive at these levels just doesn't align with the comprehensive macro data or financial market action.
Thread.
@BickerinBrattle This is such a stupid take man. You're trying so hard to push an overly simplistic narrative. Also, there's little-to-no chance Fed cuts in July.
Put this in the “I thought I’d seen everything” department: the 13 week correlation between the S&P 500 cap-weighted index and the S&P 500 equal-weighted index is now 36%. That’s between two indices that have exactly the same constituents. It’s normally 99%, as one would expect🧵
Six endpoints for a terrorist group:
1. Achieves core objectives (least likely)
2. Morphs into an insurgency
3. Repressed militarily (Israel's hope)
4. Decapitation of its leaders
5. Negotiation (anathemetic to Israel)
6. Failure by implosion or loss of support (most likely)
There is no question that eliminating Hamas would be good for Palestinians, Israel, the Middle East, and the United States, writes @akcronin. But to do so, Israel must aim to end the group not through military defeat, but instead through strategic failure. https://t.co/UlrDWSC3Qf
Many observers are coming to assume that deglobalization is a long-term certainty—but a closer look at the economic data shows that the world economy is still evolving to become more, not less, globalized in key ways, writes @Brad_Setser. https://t.co/ASQvkQ4Y8I
.@aa51_ansari examines the roots of Iran’s complex dynamic with Israel—and how the long history of Persian and Jewish coexistence could play a role in shaping the future of the Middle East:
https://t.co/odGESj7hfX
@AndrewCurran_ Note that the study's limited to music in positions 4-10 of the Top 10 Billboard for each year. It's possible that the type of music most likely to make into that top 10 status has decreased in quality, while not being predictive of the quality of music more broadly.
@AndrewCurran_ https://t.co/XiMQ9wEM8M
According to the same study, the average person born >1990 has a peak preference SSA < 0, supporting the counter-argument that music quality has indeed worsened over the years.
LoRA Land
310 Fine-tuned LLMs that Rival GPT-4
Low Rank Adaptation (LoRA) has emerged as one of the most widely adopted methods for Parameter Efficient Fine-Tuning (PEFT) of Large Language Models (LLMs). LoRA reduces the number of trainable parameters and memory usage while
I also sat down this weekend to summarize the latest LLM releases. What a month! We had four major open LLM releases: Mixtral, Meta AI's Llama 3, Microsoft's Phi-3, and Apple's OpenELM. In my new article (https://t.co/pulRbVznzy), I review and discuss all four of these major transformer-based LLM model releases, followed by new research on reinforcement learning with human feedback methods for instruction finetuning using PPO and DPO algorithms.
In short, the topics I am covering are:
1. How Good are Mixtral, Llama 3, and Phi-3?
2. OpenELM: An Efficient Language Model Family with Open-source Training and Inference Framework
3. Is DPO Superior to PPO for LLM Alignment? A Comprehensive Study
Thrilled to announce AlphaFold 3 which can predict the structures and interactions of nearly all of life’s molecules with state-of-the-art accuracy including proteins, DNA and RNA. Biology is a complex dynamical system so modeling interactions is crucial https://t.co/Gs4GoOB3fD
I am so excited that xLSTM is out. LSTM is close to my heart - for more than 30 years now. With xLSTM we close the gap to existing state-of-the-art LLMs. With NXAI we have started to build our own European LLMs. I am very proud of my team. https://t.co/IH7giCe3gd