Xavier
@xavi3rlu
Followers
2K
Following
5K
Media
64
Statuses
308
@latentholdings https://t.co/HQ6udCXyIN // prev research @opentensor quant ⇢ fundamental @panteracapital
Joined March 2021
Shining a light on three #Bittensor subnets that are driving cutting-edge research, all while being tested against rigorous scientific benchmarks. This is the innovation @opentensor loves to see 👇🧵 (0/7).
9
58
216
life update: i recently left @panteracapital and joined @opentensor research team . excited to dedicate all my time pushin' τ with @const_reborn and @shibshib89
18
18
179
TAO reserves across SNs have dipped below the amount injected by the protocol. This could be bc the continued staking pressure didn't keep up with the (slowly diminishing) root selling & faded away over the weekend. With that being said, public beta is now
18
23
167
dTAO staking activity is starting to pick up. today we saw that there is more TAO staked compared to the TAO liquidity injected by the protocol
12
29
145
dTAO cumulative trading volume reached ~$1.1bn in its first month; for reference, hyperliquid did $1.3bn in the same timeframe after launch. source: #bittensor $TAO
9
37
127
we are only <1wk into dTAO and have already seen some insane price action, but keep in mind that root proportion is still massively over alpha proportion, averaging 95% vs. 5%. we ran a simple simulation to estimate how much TAO is needed to keep dTAO prices at current levels. as
13
14
115
gonna tell my kids I worked on this.
dTAO Whitepaper First Look. Read it here: Full Conversation :: Novelty Search E031
3
0
99
With dTAO, #Bittensor will be an AI launchpad with pumpdotfun returns.
Bittensor dTAO is dropping in mid-February. Huge congratulations to @const_reborn and team @opentensor — we @latentholdings are also very proud to play a role in shipping this for everyone. 🤝🔥🌊
1
12
96
Another reason to bet on #TAO: devs can prep for a half marathon in 4 weeks during one of the most pivotal launches in #Bittensor history, and finish it in 2hr. couldve been speedier if I wasn’t providing shareholders value
7
1
88
"deploying more capiτal - sτeady lads".porτfolio viewer now live on
10
10
79
#FridayNightShip: Cross-Subnet Swap with live balance updates & dynamic fee estimates now available on 🚢
10
16
77
#Bittensor subnets should focus on two key objectives in order:. 1st: Optimize quantifiable metrics that exceed current benchmarks and advance the SoTA in research.2nd: Once this foundation is established, develop a viable, profitable product that attracts paying customers in the.
@degentradingLSD The thing that wasnt addressed which needs to be addressed is whos paying to use it?. You have to both beat the competition then you have to have enough people paying for it to be profitable. Chatgpt is burning billions a year just to try win and get better and it not profitable.
6
7
75
now you can see how fast we cycle between "we are so back" to "it's so over" on
One thing I’ve been working on lately was a DTAO Fear & Greed Index. Since the launch of DTAO, the index shows a clear swing from fear in March to greed by mid-April, with a recent move back toward neutral. Chart below. Do we want to add this to
5
7
70
TAO reserves spiked vs. protocol-injected TAO on May 2, with ~25K TAO in net inflows across all subnets. @rayon_labs's SNs saw the highest inflows, while SN55 Precog & SN62 AgenTAO led 1D perf. w/ +35% and 200 TAO in net inflows. Track net inflows on
1
10
71
During @acm_ccs in Salt Lake City the cheapest rental car was a 15-passenger van. Naturally we took a squad of academicians from all over the world, went on a scenic drive, and #TAO-pilled everyone. #bittensor
4
6
68
Ironically, the "centralization" issue it is pointing to actually proves how well decentralization is working. The foundation's delegated TAO represents part of the community's unified decision. SN28's saga was not top-down control but decentralized consensus in action. Praise
8
2
64
SN50 has the right idea of following facebook/musicgen’s evaluation methods, but relying on private or limited reference audio datasets for their evaluation metrics opens up to lookup attack and worse insider attacks in mining. The issue is worthy of investigation but we can't.
9
6
51
Running a node in @bittensor_ offers a differentiated way to impact the evolving field of decentralized AI, but it requires careful understanding and planning. Here is a practical guide on navigating this process 👇. #bittensor .
1
8
33
@guil_lambert left the following exercise in one of his Medium post. My solution here 👇🧵. Nothing crazy, maybe it can help you with your financial engineering pset :P
5
4
33
feature of the day: you can set 69min OHLC resolution, nice. and it's super fast
3
3
34
In case you missed it, @NovaWalletApp has a browser feature where you can trade with on your mobile
5
2
29
MEV cluttering your flow? Gray or filter them out under Subnet > Transactions on
5
3
26
some loosey-goosey analysis here🥪: a lot of PNL was realized recently on SN14 TAOHash, chart might be fresher than ever. fresh chart = clean slate = potentially bullish. NFA
6
1
17
(SN31: 2/2). SN31 is benchmarked against the well-studied CIFAR-10 dataset and they have produced models that outperform most SOTA models. SN31 plans to publish this result in a conference soon. If you would like to access their paper, please contact SN31 through Discord.
1
1
15
(SN31: 1/2). SN31 (NAS Chain) incentivizes miners to discover the best neural network architectures that maximizes accuracy and minimize hardware-dependent characteristics such as # of parameters and FLOPs. The goal is to maximize the Pareto optimality between these variables.
1
0
14
(SN34: 1/3). SN34’s (@BitMindAI) goal boils down to predicting if an image is real or fake on a scale of 0 to 1. Currently, there is a lack of open benchmarks for general synthetic image detection. DeepfakeBench is specifically used for faces. SN34 uses the BLIP-2 VLM to generate
1
4
13
(SN34: 3/3). SN34 is integrating more SOTA models like FLUX.1-schnell to perfect its synthetic image detection benchmark. Plus, new dashboards and graphs will soon help miners track their performance.
1
0
10
(SN34: 2/3). SN34 uses the base model architecture from DeepfakeBench and optimizes it against the dataset above. The team also introduced a novel approach called Content Aware Model Orchestration (CAMO) that enhances this by routing images to specialized expert models, improving
1
1
11
one point i missed is that other validators started childhotkey-ing to OTF's validator after const announced the anti-ponzi validator code for SN28.
1
0
8
Wait I don’t have a coffee grinder or machine how do I consume these beans.
4
0
10
(SN40: 2/2). They evaluated the effectiveness of different text chunking solutions in a RAG setup, comparing SN40’s miners against Unstructured IO’s partitioner and AI21’s text segmentation API on the GPQA benchmark, which tests with graduate-level Google-proof questions.
1
0
8
On my flight to #AvalancheSummit, the guy is front of me is explaining Bitcoin to an old Spanish lady. The world is great.
1
0
8
(SN40: 1/2). SN40 (Chunking) incentivizes segmenting text data (with plans to expand to more media types) into the most semantically meaningful chunks to enhance performance in RAG pipelines. Currently, validators generate synthetic queries by selecting featured Wikipedia.
1
1
8
Then we went to the Cheesecake Factory. The Austrian minds cannot comprehend that one slice has 910+ calories.
1
0
8
SN50 has the right idea of following facebook/musicgen’s evaluation methods, but relying on private or limited reference audio datasets for their evaluation metrics opens up to lookup attack and worse insider attacks in mining. The issue is worthy of investigation but we can't.
0
0
4
@btcrenaissance @mogmachine @KeithSingery Fixed. Now price sum excludes SNs with no pending emission of alpha or TAO, as they show no activity.
2
0
6
@mogmachine @KeithSingery @taostats the data is after all the considerations you mentioned, thanks to the ex-OTF chads who wrote the subtensor at @latentholdings . with that being said, it's always a good thing to have multiple sources. excited to see taostats' new dashboard.
1
0
6
2/9.Quick overview: leverage trading is a double-edged sword. It allows traders to amplify their exposure to an asset with borrowed funds, increasing potential P&L. Liquidation occurs when a trader's losses deplete his collateral, triggering an automatic closure of their position.
1
0
6
@tradeparadigm 🙏🙏. and shamelessly shilling my tweet .
Knowing other traders’ liquidation levels can be advantageous, similar to knowing other’s stop losses, so how do we surface such insight on @GMX_IO. This 🧵summarizes my 2-part Medium posts where there will be more in-depth w/ full python code. 1/9.
1
0
4
Ironically, the "centralization" issue it is pointing to actually proves how well decentralization is working. The foundation's delegated TAO represents part of the community's unified decision. SN28's saga was not top-down control but decentralized consensus in action. Praise
0
0
6
This also does not include auto root selling. It’s all triggered by users.
0
0
6
it's interesting to see how the #FTX Exploiter routed through multiple wallets to get to #THORChain could there be another reason besides making tracing harder?. 0xcbce eventually bridges to Threshold's tBTC, i didnt graph everything as it is getting too big and messy . 1/2
After @THORSwap suspended swaps, FTX Exploiter started swapping $ETH for $tBTC and cross-chained to the Bitcoin network through @TheTNetwork. So far, FTX Exploiter has swapped 75,636 $ETH($124M) for BTC assets and cross-chained to the Bitcoin network.
2
0
5
1
0
5
This is just a glimpse into the world of margin trading on DEX’s and there’s a lot more to explore. Stay tuned for part 2 with full python code and more related topics. Follow me on !.
1
0
2
0
0
3