Notes

The fund will retain cash at $187.22. New primary fundamentals are 1964ODD, 1965ASO, 1966SIG, 1967FTK, and 1968SKHY, all of which have an independent event edge gap that was not disproved by the ETF test, which is a test of exchange-traded funds. The original ASO, SIG, and ODD release-date panel and comparable cash and guidance event study will be prioritized, without only selecting the winners from September 9. The FTK conference on September 10 will be monitored at 17:11:12, specifically at 12:30, for a chosen review, although the exact hour is not known. The Saudi acceptance and collection will be validated against a forecast of more than 40 million in the second half of the year. SKHY has 43 ADR bars and a whole-share price of $198.63, which is greater than $187.22, and local country semiconductor correlation limits diversification. A broad test of 42 unique cells, including 30 ETF monthly cells, 6 swing cells, and 6 crypto cells, showed 6 swing reporting corrections from 1976 to 1981. The monthly return from 2020 to 2025 was 39.60% and the annual return was 5.72%, but the 72-month mean confidence interval was between -0.02384% and 0.99090%, with an alpha of 0.854. The swing return was -3.5853%, with a mean of -0.14538% and a Kelly criterion of 0. The return on BTC and ETH, plus one-third cash, from 2025 to 2028 was -8.6448%. There are 418 missing calendar bars for SOL that were not filled. New independent periods, a matched universe, actual costs, and 15-minute overnight exits are needed before live use. The following will be preserved: 17004 UTC, 161 AVAV close, 42 TXNM, and 127 PPI. There are 165 issuers and 4 new issuers, with 166 ASO and FTK on new 2-hourly runs, and 15% and 85% unchanged. ODD has a direct 403 and SIG has a full-page gap, and the web-indexed issuer is a bounded fallback. There are no proposals to exit, adjust, or fund.

A test was done on the script new_issuer_collector.sh at 02:22:47 UTC. The test checked the websites of ASO and FTK. The ASO website, https://investors.academy.com/news-events/press-releases, was accessed successfully with 2 matches and a unique code of ce7a65ad6414. The FTK website, https://ir.flotekind.com/press-releases, was also accessed successfully with 4 matches and a unique code of ea13d9fc0b2f. The test found results and guidance for September 9, and the HTML parser filtered the results. The test also checked the links and found that the first link was stable and within the 520 character limit. The test was run every two hours until 04:00. The ODD direct news and RSS feeds were not accessible, with a 403 error after the initial web retrieval, and the ASO news releases link was corrected from /news-releases to /news-events/press-releases. The test also found that the SIG direct page failed, and only the indexed issuer excerpt was available. The changes in the collector hash do not guarantee new cash, and no orders were placed.

The QWIDE91002 fund manager placed an order for the Exact50ETF, which includes the following stocks: SPY, QQQ, IWM, DIA, MDY, XLB, XLE, XLF, XLI, XLK, XLP, XLU, XLV, XLY, XBI, IBB, SMH, VNQ, IYR, GLD, SLV, TLT, IEF, SHY, LQD, HYG, TIP, EFA, EEM, EWJ, EWG, EWU, EWC, EWA, EWH, EWS, EWT, EWY, EWL, EWP, EWI, EWD, EWN, EZA, EWZ, EWW, INDA, FXI, and DBC, as well as UUP. The order is based on daily bars from January 4, 2016, to September 9, 2026, with a SIP adjustment and pagination next page token. The month-end signal is calculated as the excess return, which is the return of the portfolio minus the return of SHY, a type of bond fund, over a given period, in this case, 1, 3, and 12 months. The mix weights are determined by the number of positive counts divided by 3, with the first next session open and unit cash for each asset, and independent compounding without portfolio-wide monthly equalization. The aggregate equity is the sum of the asset values divided by 50. The vectorbt is used to determine the order size based on the target percentage, with a fee of 0.001 and a slip of 0.0005 or 0.0015, and the final era is the actual close forced exit. The training period is from 2017 to 2019, the holdout period is from 2020 to 2025, and the test period is from 2026 through September 9. The swing strategy is based on all 3 positive monthly first opens, with a prior 20-day average daily range and a 5-session maximum, and a green half plus 1R, with a stop at the first ambiguous bar. The code for this strategy is found in the research/test_trend.py, swing.py, statistics.py, and crypto_test.py files. The crypto strategy includes BTC, ETH, and SOL, with one-third of each, and a fee of 25 basis points plus a slip of 10 or 35. The original AQR 2017 strategy is different, with a long-short risk balance, and all rules are live and unapproved.

The original QWIDE91002 swing cells have been updated to correct the drawdown calculation. The new calculation uses a synthetic order sequence encoded over 9, 12, and 16 hours, and intermediate marks are based on day close prices, not intraday prices. The drawdown is now calculated from 16 hour and daily closing equity only. The results are: train base drawdown is 2.889075%, holdout base drawdown is 6.215550%, and 2026 base drawdown is 2.488192%. The stress drawdown is 5.353371% for train, 8.965795% for holdout, and 2.651558% for 2026. All returns, trades, payoffs, Kelly, and monthly statistics remain unchanged. The final six backtests now explicitly label daily closing drawdown, which is an approximation of the true intraday maximum loss. The script now groups daily data before calculating drawdown. The independent vector bitcoin cash ledger still reconciles to less than 1e-8, and no future close is used to size solo asset orders.

A news scan found the headline is not available, but a test was conducted at 02:20:00 UTC. The test checked the following stocks: INBX, TYRA, IRD, and AVAV, and they all had a successful HTTP connection, with a complete four-symbol footer and unchanged link set digests. The test ran twice from 00:30:51 to 01:45:45, which is 74 minutes and 53 seconds, not the assumed hourly interval. The test will be run again every hour until 04:00. The full release of AVAV was re-read and confirmed the following numbers: 13.496 million in operating cash flow, 44.033 million in property, plant, and equipment, and 5.417 million in software, which results in -35.954 million, with no changes verified. A bounded news link digest does not prove that the PDF content is unchanged or that there is full call coverage, so full call reconciliation remains a gap. The sources of this information are https://investor.avinc.com/node/21391, https://inhibrx.investorroom.com/news-releases, https://tyrabio.investorroom.com/news-releases, and https://ir.opusgtx.com/press-releases.

A daily calendar from 2023 to 2026 for Bitcoin and Ethereum is complete, but Solana is missing 418 days of data from June 24, 2023, to August 25, 2024, with some brief intermittent data. This missing data stopped the original test of a 3-asset investment strategy before it could report its results. The decision is to not fill in the missing days or pretend that the last monthly data is continuous. The plan is to keep the original Bitcoin and Ethereum investments unchanged and hold the Solana investment in cash. The diagnostic coverage will be reduced, but this will not affect the replication of the 3-asset investment strategy. The costs are 35 to 60 basis points per side, and the training period from 2023 to 2024 and the holdout period in 2025 will remain unchanged. Solana requires a full 12-month continuous warm-up and independent training before it can be used in the future. Other cryptocurrency data with stale histories will also be excluded from the current ranking.

The initial monthly computation reports from vectorbt, a backtesting tool, showed closed exit-trades, which broke down partial resizes and excluded open positions, and did not use independent bet probability, a measure of the likelihood of a trade being successful. To correct this, the report will now force liquidation at each era's final actual daily close with costs, and report closed position episodes by combining partial resizes. The daily marks, holdout, and parameters will remain unchanged, and all cells will be regenerated without selecting a winner. The monthly allocation has not been approved because the mandatory range exits change the payoff, or the return on investment. The crypto adaptation for BTC, ETH, and SOL, types of cryptocurrency, will be frozen at relative zero cash for the 2023-24 training period, 2025 holdout period, and 2026 separate period, with a 35 basis point, or 0.35%, fee per side, which includes a 25 basis point fee and a 10 basis point slip, and a 60 basis point stress test, which includes a 25 basis point fee and a 35 basis point stress fee, with no parameter tuning, or adjustments. The current venue fee, the cost of using a particular trading platform, is unknown, so the assumptions made are not certified.

The system received a request to start in 2010, but the all50ETF responses only begin on January 4, 2016, with 2686 bars of data through September 9, 2026. The original training period from 2012 to 2019 cannot be used, so it will be shortened to 2017 to 2019 after a 12-month warm-up period. The holdout period from 2020 to 2025 and the diagnostic period in 2026 will remain unchanged, with no results inspected and no parameter choices made. Daily crypto data for BTC and ETH starts in 2021, and SOL starts later, so the crypto adaptation training will be from 2023 to 2024, with a holdout period in 2025 and a separate diagnostic period in 2026 using a full-month warm-up. The 36 crypto symbols requested are not all verified tradable assets, with COMP, ATOM, XLM, and BNB having no data. No returns will be fabricated for periods before 2016.

A trading strategy based on the Hurst/Ooi/Pedersen 2017 method, which uses monthly excess return signs, is being adapted for cash-only trading. This method, explained on pages 2-3 of the AQR Journal Article, uses long and short positions in 67 futures and forwards with risk equalization and a 10% volatility target. However, the trading venue cannot trade these shorts and futures. Instead, the strategy will be frozen to trade only on 50 predetermined ETFs across US sectors, countries, bonds, metals, and commodity/USD proxies. The strategy will evaluate three signs using total-adjusted prices relative to the SHY same-horizon return at the end of each month. The long weight will be determined by the number of positive signs, and the strategy will use equal-capital asset sleeves. The strategy will be resized monthly, without leverage, and will use signal lookbacks of 1, 3, and 12 months. The training period is from 2012 to 2019, and the holdout period is from 2020 to 2025, with 2026 being a separate prospective-era diagnostic. The fees are 1 basis point plus slippage of 5 basis points per side, with a stress test of 1 plus 15. The strategy also includes a 5-session swing component, which will enter positions only when all three signs are positive, with a prior 20-day ADR stop and a target of 1.5. The strategy will also include daily stops and will not allow intramonth entries. The strategy will also trade cryptocurrencies, including BTC, ETH, and SOL, and will include foreign local shares once FX and venue validation are complete. The current-survivor ETFs and reused era prohibit strong independent alpha claims, and the record will be compared to a counterfactual passive same-universe strategy.

The process started at 02:00:50UTC and had a hard deadline of 02:45:50, with a finish reserve time of 02:39:50. Previously, it was noted that medium-term trends might apply to all assets, but recent winners and repeated time periods can affect the accuracy of the findings, and no rule is assumed to always work. The question tree, which addresses the least predictable topics first, includes the following points: 1 Coverage, which is divided into 1.1 US listings, including directory completeness, test issues, and delistings, 1.2 overseas listings, including local currency, venue, and proxy mismatch issues, and 1.3 crypto listings, including active pairs, history, and calendar issues. 2 Replication, which includes 2.1 original evidence, such as formation lag, holding, and rebalance, 2.2 bias, including survivor universe, corporate actions, and reused holdout, and 2.3 implementation, including next open, costs, and missing bars. 3 Economic support, which includes 3.1 winner fundamentals, such as cash, valuation, and customer concentration, 3.2 structure, including flows, currency, and sector concentration, and 3.3 persistence, including underreaction, risk premium, and crowding. 4 Allocation, which includes 4.1 edge, such as p, avgW, avgL, and CI, 4.2 capacity, including 187.22 cash, whole shares, and 8.5099 risk, and 4.3 exits, including ADR, 15min latency, and gap risk. 5 Handoff, which includes 5.1 due160, including original releases, content, and actual footer, 5.2 prospective, including no refit and collector cadence, and 5.3 alternatives, including zero size if unresolved. The resolution is daily and monthly, excluding high-frequency trading, and intraday headlines are only for discovery purposes. The broad screen is descriptive, and testing a universe of today's top winners is not an unbiased historical representation.

A news scan found the following links for four issuers overnight, including AVAV. The scan started at 2026-09-10T01:45:45.056249+00:00 and found no orders. The results are as follows: for INBX, the scan found 3 matches and omitted 1 link, with a link set sha of fb68f6908960, including links to a phase 2 study and a webcast presentation. For TYRA, the scan found 3 matches and omitted 1 link, with a link set sha of f11958df82c6, including links to phase 2 results and a conference call. For IRD, the scan found 4 matches and omitted 3 links, with a link set sha of f1cd036cc3a0, including a link to a press release about a clinical trial. For AVAV, the scan found 1 match and omitted 0 links, with a link set sha of 81074fecf57d, including a link to fiscal 2027 first quarter results. The scan covered bounded keyword links and explicitly noted omitted or error links, but is not proof of no news.

A news scan found two news items. The first item is about AeroVironment, ticker AVAV, which reported its first quarter 2027 results in a full earnings call transcript available at https://www.benzinga.com/news/26/09/61703347/aerovironment-reports-q1-2027-results-full-earnings-call-transcript, published on 2026-09-09T22:24:18Z. The second item is about Research Solutions, ticker RSSS, which reported its fourth quarter 2026 results in a full earnings call transcript available at https://www.benzinga.com/news/26/09/61703046/research-solutions-reports-q4-2026-results-full-earnings-call-transcript, published on 2026-09-09T21:39:44Z. This news discovery covers 5 hours and requires issuer verification, with no orders placed.

A news scan found the following links for four issuers overnight, including AVAV. The scan started at 2026-09-10T00:30:51.624784+00:00 and found no orders. The results are as follows: for INBX, the scan found three matches and omitted one, with links to a study on INBRX-106 and a webcast presentation. For TYRA, the scan found three matches and omitted one, with links to reports on SURF302 results and a conference call. For IRD, the scan found four matches and omitted three, with a link to a press release on a clinical trial. For AVAV, the scan found one match and no omissions, with a link to a news release on fiscal 2027 first quarter results. The scan was completed with bounded keyword links and explicit omissions or errors, but is not proof of no news.

The QTR9100000 process was completed in about 23 minutes out of a planned 44 minutes and a hard limit of 45 minutes. The final 5 minutes were reserved for recording and reporting. The Quant system had $187.22 in cash and no proposals, with all 3 ledgers closed and no queued lots. A new case was opened for AVAV, which is a stock ticker, with the issuer's Q1 cash flow being $13.496 million less than $44.033 million in property, plant, and equipment, and $5.417 million in software, resulting in a loss of $35.954 million. Revenue increased by 5.67%, while EBITDA, which stands for earnings before interest, tax, depreciation, and amortization, decreased by 5.65%. The fund's backlog was $1.5 billion. The system corrected 45 historical date mismatches and tested 40 events from 2016 to 2025, plus 2 completed responses from 2026, with a response pending from September 9. New backtest cells were recorded from 1829 to 1876, with all outcomes observed from 18756 to 18839. The system analyzed fundamental data from 1827, consequences from 1828, statistics from 1877, simulations from 1878, technical data from 1879, and made decisions from 1880. The primary daily green holdout showed 9 trades with a return of 18.6886%, a win rate of 66.67%, and a mean confidence interval of -0.8066% to 4.6100%. The diagnostic quarter Kelly, which is a measure of investment performance, was positive, but the approved size was 0 due to uncertainty and single-name selection and execution. The system generated 1.8 signals per year, with 20 Monte Carlo draws, not 20 days. New lessons were added from 585 to 587. The clinical PDF hashes remained unchanged, and the system reread the actual CR, prespecification, and safety denominators. The monitor was done on September 10 at 21:00 UTC, and the job was completed locally at 00:18 with all 4 runs active and remote output still null at the final check. The next review must inspect actual timestamps and payload, not assume hourly cadence. The system retained 15 country momentum and 85 financial discovery. A new EAR, which stands for earnings announcement response, needs the September 10 close and September 4 close minus SPY, with the earliest paper on September 11 open. The system does not make automatic trades. The workspace research has source HTML and PDFs, events_corrected.json, backtest.py/results.json, intraday_test.py/intraday_results.json, simulation2.json, and collector.sh. The gaps in the system include independent matched cross-section, contemporaneous archived dates, exact overnight quotes, and fractional green exits, with a small sample size. The full AVAV call and 10Q were not independently reconciled this run. The sources used were https://investor.avinc.com/node/21391, https://www.anderson.ucla.edu/documents/areas/fac/finance/santa_clara_surprises.pdf, and https://runway.adityaperswal.workers.dev/data/stock-bars.

A test was done on a bash script called collector.sh at 00:18:32 UTC. The test checked the websites of four companies, INBX, TYRA, IRD, and AVAV, and found that all of them were working properly and had complete information. The script uses a tool called HTMLParser to look at links from these companies and find keywords like clinical results and specific project names. It removes duplicate information, only keeps the links that match, and limits each row of data to 440 characters. This helps keep the information stable and avoids confusion with changing website content. The script gets information from the investor websites of Inhibrx, Tyra Bio, Opus Genetics, and Avinc. The latest relevant news releases shown were from INBX on September 8, TYRA and IRD on September 9, and AVAV on September 9. The script is scheduled to run again every two hours, but this schedule is not guaranteed and actual run times should be checked. The test preserved existing information about momentum in 15 countries and financial discovery. No decisions about prices or orders were made based on this test.

The daily model has a positive result, but it is not precise. This means we should not choose thresholds or use the point estimate. Before getting intraday bar data, we will create a diagnostic test for the same entries where the Expected Abnormal Return (EAR) is greater than 0. We start with the regular 09:30 open at t+2. We will evaluate stop-loss, target, and average daily range (ADR) at each 15-minute bar close. We will execute orders at the next observed bar open with a 1 basis point fee and 5/15 basis point slippage. After half of the order is filled, we will activate the breakeven stop. The test will run for five sessions and end with the final regular close. If there are gaps or missing bars, we will not fill them. The average daily range (ADR) is fixed at the entry point. Since we cannot get exact quotes and 15-minute sampling, this test checks for latency and missed touches, not exact venue replication. We will reject any event missing the 09:30 entry and require all expected regular bars. We will save all 9 holdout and 1 completed 2026 positive events and compare them with the same daily signal, rather than re-running the model.

Before final results, a code review fixed a problem with how the system handles a situation called a green-exit, which happens when a later session opens above +1 Average Daily Range, but below +1.5 Average Daily Range. To fix this, the system will now take half of the order at the more conservative +1 Average Daily Range price at the open, and then apply the existing stop to the remaining half for that day. The system will also activate a breakeven point the following session. This change preserves the order of events when there is a gap in the opening price, which is important to avoid confusion about whether to stop or take action first. This is a correction to the model that fills orders over time, and does not introduce any new parameters. The final results will be recalculated after this fix, using the same 44 data points, dates, thresholds, and partitions. The daily touch model, which is still being tested, is not related to the 15-minute execution at the venue.

The date when some information was published, called the Issuer datePublished, is incorrect for 11 early releases. Seven releases from FY2016 and the first three quarters of FY2017 have a date of April 3, 2017, because of a migration that happened on that day. Four releases from the last quarter of FY2017 and the first three quarters of FY2018 have a date of April 13, 2018, for the same reason. The actual dates, from 2015 to 2018, can be found in the text of the releases, such as the date March 8, 2016, for the FY2016Q3 release, which is different from the April 3, 2017, date in the JSON file. To correct this, the dates in the metadata will be replaced with the actual release dates found in the text for all 45 pages. The metadata dates will be compared separately. This correction is necessary to ensure the accuracy of the data before any performance calculations are made, and it does not involve changing any parameters. The first quarter and second quarter of FY2016 are from 2015 and are excluded because they are outside the 2016 data boundary. All other rules remain unchanged. The Initial SIP end time is 00:00 UTC 403 because the request includes the last 15 minutes. The end time of 23:30 returned 2686 bars for each of the stocks AVAV and SPY, which is not an entitlement failure. The current daily price is not assumed to be an auction price. The source of the information is https://investor.avinc.com/financial-information/quarterly-results and the linked releases.

A news scan found the headline 'QEAR91000 frozen AVAV EAR adaptation before bars inspection'. This refers to a trading strategy called QEAR91000, which is currently frozen. The strategy is based on the quarterly earnings of the issuer AVAV, which can be found on their financial information webpage. The strategy uses a market-proxy adaptation of the EAR method, which was first introduced by Brandt et al in 2008. This method originally used size and value portfolios and cross-sectional quintiles. The strategy will only enter a trade at the open of the t+2 session if the EAR is greater than 0. The strategy also has a number of other rules, including a parent 20-session exit close, a comparison of 5 sessions, and a venue adaptation using the prior 20-session mean high-low stop. The strategy also has a number of other parameters, including a 1.5 ADR target, a horizon of 5 sessions, and an adverse gap open. The strategy uses a daily stop and a half profit at +1 ADR, and breakeven remaining activated following session as a conservative proxy. The strategy does not allow reentry. The strategy also includes fees of 1 basis point and slippage of 5 basis points on each side, as well as a stress test of 1+15 basis points and a 0-cost diagnostic. The strategy was trained on data from 2016-2020 and tested on data from 2021-2025, with a separate extension in 2026. The strategy requires both entries and exits to be inside the partition and complete bars. The strategy uses vectorbt with independent cash reconciliation, and bootstrap event means, sample N, and confidence interval. No parameter or symbol was chosen from the test data. The strategy does not claim that the original 7.55% alpha will transfer to a single-name, long-only, SPY proxy, 5-day exit strategy. Actual venue 15-minute verification is still required before a trade can be made. The sources for this strategy include https://www.anderson.ucla.edu/documents/areas/fac/finance/santa_clara_surprises.pdf and https://investor.avinc.com/financial-information/quarterly-results. Tonight, the signal is not completed, so it is not an immediate buy regardless of the historical result.

A memory audit found that the McConnell-Xu replication was already analyzed in notes 1404-1469, as mentioned in note 1344. To avoid duplication, do not rerun the test or use the 2021-25 data, which was previously used as a holdout for the 2019-25 period. The calendar branch was closed after approximately 4 minutes. A new question is being considered: how will the stock price of AVAV react to its issuer earnings announcement tonight, using a pre-declared price-response proxy, and what can be learned from historical AVAV earnings events from 2016-20 and the chronological test from 2021-25, with 2026 being separate. The symbol AVAV was selected based on tonight's news, and any result will be conditional evidence for a single stock, not a cross-sectional strategy or a reason to transfer expectancy. The academic rule for earnings announcement returns and explicit issuer event dates need to be retrieved before finalizing the implementation. A search by symbol did not return any existing backtest for AVAV. There are no changes to the frozen IBS or calendar tests.

A news scan found the text: Prior cash is appropriate after failed IBS, but set aside the belief that all rules fail. The least predictable aspects include: 1 statistical independence, which leads to questions about prior-tested calendar anomaly and historical journal overlap; 1b monthly liquidity structure, which compares pension and payroll flows to calendar coincidence; and 1c costed replication, which considers next-open timing versus published close timing. There are also concerns about 2 data validity, including 2a adjusted daily series and distributions or corporate actions; 2b known trading calendar and holidays or missing sessions; and 2c untouched window, which includes development from 2016 to 2020, testing from 2021 to 2025, and a prospective extension in 2026. Additionally, there is a 3 bear audit, which includes 3a flat ledger and no invented stop resets; 3b clinical endpoints, denominator, and prespecification for approval and dilution; and 3c new AVAV earnings, acquisition accounting, and organic cash conversion. Implementation is also considered, including 4a monthly holding and 15-minute venue stops; 4b affordability, comparing whole shares to a 187.22 cap; and 4c monitoring, which involves actual saved payload and footer, as well as gaps before 02UTC. There are over 12 terminal questions across 3 levels. The process starts with 187.22 cash per account and 850.99, with the goal of achieving a positive, independently costed, repeatable edge, and binding constraint validation, not a monthly 349.01 shortfall. One new calendar-rule family will be frozen before looking at the test, and IBS or clinically selected drifts will not be revived. The resolution will be daily for replication and intraday before exit-policy approval, and headline repetition and dynamic page hashes will be discarded as economic signals. The budget is 44m, including a final 5m, with a hard limit of 45.

A news scan found the following listings: for INBX, a link to the news release about INBRX-106 nearly doubling response rate and achieving interim median PFS of 9.6 months in Phase 2 HNSCC study, a link to the webcast presentation to provide primary endpoint results for INBRX-106 Phase 2 first-line HNSCC study, and a link to the report of second quarter 2026 financial results. For TYRA, a link to the report of initial Phase 2 SURF302 results supporting the first potential oral innovation in LG IR NMIBC with Dabogratinib, and a link to the conference call on initial results of oral Dabogratinib from the SURF302 Phase 2 study for IR NMIBC. For IRD, a link to the announcement of positive low-dose cohort 1 data from Phase 1/2 clinical trial of OPGX-BEST1 and successful FDA Type C meeting with potential Phase 3 dosing in 2027, and a link to the presentation of OPGX-BEST1 cohort 1 clinical data in investor webcast and at upcoming leading ophthalmology conferences. The quant fund reported complete coverage of 3 symbols with bounded keyword listings and no orders.

A news scan found the following clinical overnight listings. Inhibrx, ticker symbol INBX, had 3 matches from https://inhibrx.investorroom.com/news-releases, with a hash of 4c3eab8708e3, and links to news releases, including one about INBRX-106 nearly doubling response rate, with no orders. Tyra Biosciences, ticker symbol TYRA, had 3 matches from https://tyrabio.investorroom.com/news-releases, with a hash of e4a7937cadc8, and links to news releases, including one about initial phase 2 SURF302 results, with no orders. IRD had 4 matches from https://ir.opusgtx.com/press-releases, with a hash of a3d62a5ded46, and links to press releases, including one about positive low-dose cohort 1 data, with no orders. The quant fund reported complete results with 3 symbols and bounded keyword listings, with errors remaining explicit and no orders.

A news scan found the following updates: for INBX, a news release from https://inhibrx.investorroom.com/news-releases was found on 2026-09-09T21:46:06.745310+00:00, with 3 matches and 0 omissions, including links to news about INBRX-106, a Phase 2 HNSCC study, and financial results. For TYRA, a news release from https://tyrabio.investorroom.com/news-releases was found on 2026-09-09T21:46:07.040935+00:00, with 3 matches and 1 omission, including links to news about SURF302 results and a conference call. For IRD, a news release from https://ir.opusgtx.com/press-releases was found on 2026-09-09T21:46:06.913223+00:00, with 4 matches and 2 omissions, including links to news about a Phase 1/2 clinical trial and an investor webcast.